CoRL
Collection
Accepted papers for CoRL (Conference on Robot Learning), one dataset per year. • 10 items • Updated
paper_id stringlengths 10 10 | title stringlengths 25 127 | paper_url stringlengths 42 42 | authors listlengths 2 22 | type stringclasses 0
values | primary_area stringclasses 0
values | abstract large_stringlengths 653 2.03k | keywords listlengths 1 8 | TL;DR large_stringlengths 40 250 ⌀ | submission_number int64 7 728 | arxiv_id stringlengths 10 10 ⌀ | arxiv_id_source stringclasses 2
values |
|---|---|---|---|---|---|---|---|---|---|---|---|
WnSl42M9Z4 | HumanPlus: Humanoid Shadowing and Imitation from Humans | https://openreview.net/forum?id=WnSl42M9Z4 | [
"Zipeng Fu",
"Qingqing Zhao",
"Qi Wu",
"Gordon Wetzstein",
"Chelsea Finn"
] | null | null | One of the key arguments for building robots that have similar form factors to human beings is that we can leverage the massive human data for training.Yet, doing so has remained challenging in practice due to the complexities in humanoid perception and control, lingering physical gaps between humanoids and humans in m... | [
"Humanoids",
"Learning from Human Data",
"Whole-Body Control"
] | a full-stack humanoid system for learning motion and autonomous skills from human data | 7 | 2406.10454 | title_snapshot |
jart4nhCQr | Learning to Manipulate Anywhere: A Visual Generalizable Framework For Reinforcement Learning | https://openreview.net/forum?id=jart4nhCQr | [
"Zhecheng Yuan",
"Tianming Wei",
"Shuiqi Cheng",
"Gu Zhang",
"Yuanpei Chen",
"Huazhe Xu"
] | null | null | Can we endow visuomotor robots with generalization capabilities to operate in diverse open-world scenarios? In this paper, we propose Maniwhere, a generalizable framework tailored for visual reinforcement learning, enabling the trained robot policies to generalize across a combination of multiple visual disturbance typ... | [
"Visual Generalization",
"Sim2real",
"Reinforcement Learning"
] | A generalizable framework for visual RL, and can facilitate the sim2real transfer. | 8 | 2407.15815 | title_snapshot |
LZh48DTg71 | Evaluating Real-World Robot Manipulation Policies in Simulation | https://openreview.net/forum?id=LZh48DTg71 | [
"Xuanlin Li",
"Kyle Hsu",
"Jiayuan Gu",
"Oier Mees",
"Karl Pertsch",
"Homer Rich Walke",
"Chuyuan Fu",
"Ishikaa Lunawat",
"Isabel Sieh",
"Sean Kirmani",
"Sergey Levine",
"Jiajun Wu",
"Chelsea Finn",
"Hao Su",
"Quan Vuong",
"Ted Xiao"
] | null | null | The field of robotics has made significant advances towards generalist robot manipulation policies. However, real-world evaluation of such policies is not scalable and faces reproducibility challenges, issues that are likely to worsen as policies broaden the spectrum of tasks they can perform. In this work, we demonstr... | [
"real-to-sim",
"policy evaluation",
"robot manipulation"
] | We introduce a suite of simulated environments for scalable evaluation of real-robot generalist policies, and show strong correlation to real robot evaluations. | 9 | 2405.05941 | title_snapshot |
cT2N3p1AcE | Visual Whole-Body Control for Legged Loco-Manipulation | https://openreview.net/forum?id=cT2N3p1AcE | [
"Minghuan Liu",
"Zixuan Chen",
"Xuxin Cheng",
"Yandong Ji",
"Ri-Zhao Qiu",
"Ruihan Yang",
"Xiaolong Wang"
] | null | null | We study the problem of mobile manipulation using legged robots equipped with an arm, namely legged loco-manipulation. The robot legs, while usually utilized for mobility, offer an opportunity to amplify the manipulation capabilities by conducting whole-body control. That is, the robot can control the legs and the arm ... | [
"Robot Learning; Reinforcement Learning; Imitation Learning; Mobile Loco-Manipulation"
] | null | 12 | 2403.16967 | title_snapshot |
7ddT4eklmQ | ACE: A Cross-platform and visual-Exoskeletons System for Low-Cost Dexterous Teleoperation | https://openreview.net/forum?id=7ddT4eklmQ | [
"Shiqi Yang",
"Minghuan Liu",
"Yuzhe Qin",
"Runyu Ding",
"Jialong Li",
"Xuxin Cheng",
"Ruihan Yang",
"Sha Yi",
"Xiaolong Wang"
] | null | null | Bimanual robotic manipulation with dexterous hands has a large potential workability and a wide workspace as it follows the most natural human workflow.
Learning from human demonstrations has proven highly effective for learning a dexterous manipulation policy. To collect such data, teleoperation serves as a straightfo... | [
"Teleopration System; Hardware; Imitation Learning; Robot Learning; Exoskeletons"
] | A novel and powerful teleoperation system for dexterous tasks. | 13 | 2408.11805 | title_judge |
9XV3dBqcfe | Generalized Animal Imitator: Agile Locomotion with Versatile Motion Prior | https://openreview.net/forum?id=9XV3dBqcfe | [
"Ruihan Yang",
"Zhuoqun Chen",
"Jianhan Ma",
"Chongyi Zheng",
"Yiyu Chen",
"Quan Nguyen",
"Xiaolong Wang"
] | null | null | The agility of animals, particularly in complex activities such as running, turning, jumping, and backflipping, stands as an exemplar for robotic system design. Transferring this suite of behaviors to legged robotic systems introduces essential inquiries: How can a robot be trained to learn multiple locomotion behavior... | [
"Legged Robots",
"Imitation Learning",
"Agile Locomotion"
] | Our system learns a multiple agile locomotion skills with a single instructable motion prior from a diverse reference motion dataset. | 16 | 2310.01408 | title_snapshot |
S2Jwb0i7HN | DextrAH-G: Pixels-to-Action Dexterous Arm-Hand Grasping with Geometric Fabrics | https://openreview.net/forum?id=S2Jwb0i7HN | [
"Tyler Ga Wei Lum",
"Martin Matak",
"Viktor Makoviychuk",
"Ankur Handa",
"Arthur Allshire",
"Tucker Hermans",
"Nathan D. Ratliff",
"Karl Van Wyk"
] | null | null | A pivotal challenge in robotics is achieving fast, safe, and robust dexterous grasping across a diverse range of objects, an important goal within industrial applications. However, existing methods often have very limited speed, dexterity, and generality, along with limited or no hardware safety guarantees. In this wor... | [
"Dexterous Grasping",
"Geometric Fabrics",
"Reinforcement Learning",
"Teacher-Student Distillation",
"Sim-to-Real Transfer"
] | DextrAH-G is a safe, continuously reacting pixels-to-action policy that achieves state-of-the-art dexterous grasping in the real world and was trained entirely in simulation using RL and teacher-student distillation with a geometric fabric controller | 22 | 2407.02274 | title_snapshot |
1jc2zA5Z6J | Get a Grip: Multi-Finger Grasp Evaluation at Scale Enables Robust Sim-to-Real Transfer | https://openreview.net/forum?id=1jc2zA5Z6J | [
"Tyler Ga Wei Lum",
"Albert H. Li",
"Preston Culbertson",
"Krishnan Srinivasan",
"Aaron Ames",
"Mac Schwager",
"Jeannette Bohg"
] | null | null | This work explores conditions under which multi-finger grasping algorithms can attain robust sim-to-real transfer. While numerous large datasets facilitate learning *generative* models for multi-finger grasping at scale, reliable real-world dexterous grasping remains challenging, with most methods degrading when deploy... | [
"Multi-Fingered Grasping",
"Large-Scale Grasp Dataset",
"Sim-to-Real"
] | We show that learned grasp evaluators enable robust real-world dexterous grasping when trained at sufficient scale with perceptual data. We release a dataset of 3.5M labeled grasps and show evaluators trained on it achieve SOTA hardware performance. | 23 | 2410.23701 | title_snapshot |
pPhTsonbXq | GraspSplats: Efficient Manipulation with 3D Feature Splatting | https://openreview.net/forum?id=pPhTsonbXq | [
"Mazeyu Ji",
"Ri-Zhao Qiu",
"Xueyan Zou",
"Xiaolong Wang"
] | null | null | The ability for robots to perform efficient and zero-shot grasping of object parts is crucial for practical applications and is becoming prevalent with recent advances in Vision-Language Models (VLMs). To bridge the 2D-to-3D gap for representations to support such a capability, existing methods rely on neural fields (N... | [
"Zero-shot manipulation",
"Gaussian Splatting",
"Keypoint Tracking"
] | null | 26 | 2409.02084 | title_snapshot |
MfIUKzihC8 | CtRL-Sim: Reactive and Controllable Driving Agents with Offline Reinforcement Learning | https://openreview.net/forum?id=MfIUKzihC8 | [
"Luke Rowe",
"Roger Girgis",
"Anthony Gosselin",
"Bruno Carrez",
"Florian Golemo",
"Felix Heide",
"Liam Paull",
"Christopher Pal"
] | null | null | Evaluating autonomous vehicle stacks (AVs) in simulation typically involves replaying driving logs from real-world recorded traffic. However, agents replayed from offline data are not reactive and hard to intuitively control. Existing approaches address these challenges by proposing methods that rely on heuristics or g... | [
"offline reinforcement learning",
"autonomous driving",
"simulation"
] | We propose reactive and controllable agents in a driving simulator using offline reinforcement learning. | 27 | 2403.19918 | title_snapshot |
uJBMZ6S02T | Real-to-Sim Grasp: Rethinking the Gap between Simulation and Real World in Grasp Detection | https://openreview.net/forum?id=uJBMZ6S02T | [
"Jia-Feng Cai",
"Zibo Chen",
"Xiao-Ming Wu",
"Jian-Jian Jiang",
"Yi-Lin Wei",
"Wei-Shi Zheng"
] | null | null | For 6-DoF grasp detection, simulated data is expandable to train more powerful model, but it faces the challenge of the large gap between simulation and real world. Previous works bridge this gap with a sim-to-real way. However, this way explicitly or implicitly forces the simulated data to adapt to the noisy real data... | [
"Grasp pose detection",
"simulated datasets",
"sim-to-real"
] | A sim-to-real mehod for 6-DoF grasp poses detection | 30 | 2410.06521 | title_snapshot |
Q2lGXMZCv8 | LLARVA: Vision-Action Instruction Tuning Enhances Robot Learning | https://openreview.net/forum?id=Q2lGXMZCv8 | [
"Dantong Niu",
"Yuvan Sharma",
"Giscard Biamby",
"Jerome Quenum",
"Yutong Bai",
"Baifeng Shi",
"Trevor Darrell",
"Roei Herzig"
] | null | null | In recent years, instruction-tuned Large Multimodal Models (LMMs) have been successful at several tasks, including image captioning and visual question answering; yet leveraging these models remains an open question for robotics. Prior LMMs for robotics applications have been extensively trained on language and action ... | [
"LMMs",
"Vision Action Instruction Tuning",
"Robot Learning"
] | We propose LLARVA, a model trained with a novel instruction tuning method that leverages structured prompts to unify a range of robotic configurations and introduces the concept of visual traces to align the vision and action spaces further. | 34 | 2406.11815 | title_snapshot |
Yce2jeILGt | Open-TeleVision: Teleoperation with Immersive Active Visual Feedback | https://openreview.net/forum?id=Yce2jeILGt | [
"Xuxin Cheng",
"Jialong Li",
"Shiqi Yang",
"Ge Yang",
"Xiaolong Wang"
] | null | null | Teleoperation serves as a powerful method for collecting on-robot data essential for robot learning from demonstrations. The intuitiveness and ease of use of the teleoperation system are crucial for ensuring high-quality, diverse, and scalable data. To achieve this, we propose an immersive teleoperation system $\textbf... | [
"Teleoperation",
"VR/AR",
"Imitation Learning"
] | An immersive teleoperation system with active visual feedback | 36 | 2407.01512 | title_snapshot |
AGG1zlrrMw | Neural Attention Field: Emerging Point Relevance in 3D Scenes for One-Shot Dexterous Grasping | https://openreview.net/forum?id=AGG1zlrrMw | [
"Qianxu Wang",
"Congyue Deng",
"Tyler Ga Wei Lum",
"Yuanpei Chen",
"Yaodong Yang",
"Jeannette Bohg",
"Yixin Zhu",
"Leonidas Guibas"
] | null | null | One-shot transfer of dexterous grasps to novel scenes with object and context variations has been a challenging problem. While distilled feature fields from large vision models have enabled semantic correspondences across 3D scenes, their features are point-based and restricted to object surfaces, limiting their capabi... | [
"Desterous Grasping",
"One-Shot Manipulation",
"Distilled Feature Field",
"Neural Implicit Field",
"Self-Supervised Learning"
] | null | 42 | 2410.23039 | title_snapshot |
E4K3yLQQ7s | Visual Manipulation with Legs | https://openreview.net/forum?id=E4K3yLQQ7s | [
"Xialin He",
"Chengjing Yuan",
"Wenxuan Zhou",
"Ruihan Yang",
"David Held",
"Xiaolong Wang"
] | null | null | Animals have the ability to use their arms and legs for both locomotion and manipulation. We envision quadruped robots to have the same versatility. This work presents a system that empowers a quadruped robot to perform object interactions with its legs, drawing inspiration from non-prehensile manipulation techniques. ... | [
"Legged robots",
"Non-prehensile manipulation",
"Reinforcement Learning"
] | This work presents a system that empowers quadruped robot to perform object interactions with its legs | 44 | 2410.11345 | title_snapshot |
InT87E5sr4 | Dreamitate: Real-World Visuomotor Policy Learning via Video Generation | https://openreview.net/forum?id=InT87E5sr4 | [
"Junbang Liang",
"Ruoshi Liu",
"Ege Ozguroglu",
"Sruthi Sudhakar",
"Achal Dave",
"Pavel Tokmakov",
"Shuran Song",
"Carl Vondrick"
] | null | null | A key challenge in manipulation is learning a policy that can robustly generalize to diverse visual environments. A promising mechanism for learning robust policies is to leverage video generative models, which are pretrained on large-scale datasets of internet videos. In this paper, we propose a visuomotor policy lear... | [
"Imitation Learning",
"Visuomotor Policy",
"Video Generation"
] | We introduce a visuomotor policy based on conditional video generation and 3D tracking which is much more generalizable in manipulation tasks than traditional behavior cloning methods. | 46 | 2406.16862 | title_snapshot |
cGswIOxHcN | Learning Visual Parkour from Generated Images | https://openreview.net/forum?id=cGswIOxHcN | [
"Alan Yu",
"Ge Yang",
"Ran Choi",
"Yajvan Ravan",
"John Leonard",
"Phillip Isola"
] | null | null | Fast and accurate physics simulation is an essential component of robot learning, where robots can explore failure scenarios that are difficult to produce in the real world and learn from unlimited on-policy data. Yet, it remains challenging to incorporate RGB-color perception into the sim-to-real pipeline that matches... | [
"Generative AI",
"Simulation",
"Legged Locomotion",
"Sensory Motor-learning"
] | Train a State-of-the-art quadruped parkour policy on generated data | 47 | 2411.00083 | title_snapshot |
lpjPft4RQT | TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correction | https://openreview.net/forum?id=lpjPft4RQT | [
"Yunfan Jiang",
"Chen Wang",
"Ruohan Zhang",
"Jiajun Wu",
"Li Fei-Fei"
] | null | null | Learning in simulation and transferring the learned policy to the real world has the potential to enable generalist robots. The key challenge of this approach is to address simulation-to-reality (sim-to-real) gaps. Previous methods often require domain-specific knowledge *a priori*. We argue that a straightforward way ... | [
"Sim-to-Real Transfer",
"Human-in-the-Loop",
"Robot Manipulation"
] | We present TRANSIC, a holistic human-in-the-loop method to tackle sim-to-real transfer for contact-rich robotic arm manipulation tasks. | 52 | 2405.10315 | title_snapshot |
k4Nnxqcwt8 | Q-SLAM: Quadric Representations for Monocular SLAM | https://openreview.net/forum?id=k4Nnxqcwt8 | [
"Chensheng Peng",
"Chenfeng Xu",
"Yue Wang",
"Mingyu Ding",
"Heng Yang",
"Masayoshi Tomizuka",
"Kurt Keutzer",
"Marco Pavone",
"Wei Zhan"
] | null | null | In this paper, we reimagine volumetric representations through the lens of quadrics. We posit that rigid scene components can be effectively decomposed into quadric surfaces. Leveraging this assumption, we reshape the volumetric representations with million of cubes by several quadric planes, which results in more accu... | [
"Neural Radiance Fields",
"Simultaneous Localization and Mapping"
] | null | 53 | 2403.08125 | title_snapshot |
WLOTZHmmO6 | Let Occ Flow: Self-Supervised 3D Occupancy Flow Prediction | https://openreview.net/forum?id=WLOTZHmmO6 | [
"Yili Liu",
"Linzhan Mou",
"Xuan Yu",
"Chenrui Han",
"Sitong Mao",
"Rong Xiong",
"Yue Wang"
] | null | null | Accurate perception of the dynamic environment is a fundamental task for autonomous driving and robot systems. This paper introduces Let Occ Flow, the first self-supervised work for joint 3D occupancy and occupancy flow prediction using only camera inputs, eliminating the need for 3D annotations. Utilizing TPV for unif... | [
"3D occupancy prediction",
"occupancy flow",
"Neural Radiance Field"
] | the first self-supervised work to joint predict the 3D occupancy and occupancy flow | 56 | 2407.07587 | title_snapshot |
cq2uB30uBM | Pre-emptive Action Revision by Environmental Feedback for Embodied Instruction Following Agents | https://openreview.net/forum?id=cq2uB30uBM | [
"Jinyeon Kim",
"Cheolhong Min",
"Byeonghwi Kim",
"Jonghyun Choi"
] | null | null | When we, humans, perform a task, we consider changes in environments such as objects' arrangement due to interactions with objects and other reasons; e.g., when we find a mug to clean, if it is already clean, we skip cleaning it. But even the state-of-the-art embodied agents often ignore changed environments when perfo... | [
"Replanning",
"Environmental Feedback",
"Brain plasticity",
"Embodied AI"
] | null | 58 | null | null |
2SYFDG4WRA | Manipulate-Anything: Automating Real-World Robots using Vision-Language Models | https://openreview.net/forum?id=2SYFDG4WRA | [
"Jiafei Duan",
"Wentao Yuan",
"Wilbert Pumacay",
"Yi Ru Wang",
"Kiana Ehsani",
"Dieter Fox",
"Ranjay Krishna"
] | null | null | Large-scale endeavors like RT-1 and widespread community efforts such as Open-X-Embodiment have contributed to growing the scale of robot demonstration data. However, there is still an opportunity to improve the quality, quantity, and diversity of robot demonstration data. Although vision-language models have been show... | [
"Robot Learning; Multimodal Large Language Model; Data Generation; Imitation Learning; Behavior Cloning"
] | A scalable automated demonstration generation method leveraging on Vision-Language Models for real-world robotics manipulation | 62 | 2406.18915 | title_snapshot |
KcW31O0PtL | Hint-AD: Holistically Aligned Interpretability in End-to-End Autonomous Driving | https://openreview.net/forum?id=KcW31O0PtL | [
"Kairui Ding",
"Boyuan Chen",
"Yuchen Su",
"Huan-ang Gao",
"Bu Jin",
"Chonghao Sima",
"Xiaohui Li",
"Wuqiang Zhang",
"Paul Barsch",
"Hongyang Li",
"Hao Zhao"
] | null | null | End-to-end architectures in autonomous driving (AD) face a significant challenge in interpretability, impeding human-AI trust. Human-friendly natural language has been explored for tasks such as driving explanation and 3D captioning. However, previous works primarily focused on the paradigm of declarative interpretabil... | [
"Interpretability",
"Language alignment",
"Autonomous driving"
] | Aligning natural language to intermediate states of autonomous driving model for aligned interpretability and state-of-the-art captioning accuracy. | 67 | 2409.06702 | title_snapshot |
HlxRd529nG | Detect Everything with Few Examples | https://openreview.net/forum?id=HlxRd529nG | [
"Xinyu Zhang",
"Yuhan Liu",
"Yuting Wang",
"Abdeslam Boularias"
] | null | null | Few-shot object detection aims at detecting novel categories given only a few example images. It is a basic skill for a robot to perform tasks in open environments. Recent methods focus on finetuning strategies, with complicated procedures that prohibit a wider application. In this paper, we introduce DE-ViT, a few-sho... | [
"Robot Vision",
"Object Detection and Recognition",
"Few-shot Learning"
] | We introduce DE-ViT, a few-shot object detector without the need for finetuning, which establishes new state-of-the-art on all few-shot detection benchmarks (Pascal VOC, COCO, LVIS), and we evaluate DE-ViT with a real robot in sorting novel objects. | 70 | 2309.12969 | title_snapshot |
Yw5QGNBkEN | Scaling Manipulation Learning with Visual Kinematic Chain Prediction | https://openreview.net/forum?id=Yw5QGNBkEN | [
"Xinyu Zhang",
"Yuhan Liu",
"Haonan Chang",
"Abdeslam Boularias"
] | null | null | Learning general-purpose models from diverse datasets has achieved great success in machine learning. In robotics, however, existing methods in multi-task learning are typically constrained to a single robot and workspace, while recent work such as RT-X requires a non-trivial action normalization procedure to manually ... | [
"Multi-Task Robot Learning",
"Manipulation"
] | We propose visual kinematics transformer, a convolution-free architecture trained with forecasting kinematic structures through optimal point set matching, and demonstrate its superior performance as a general agent in diverse environments. | 71 | 2406.07837 | title_snapshot |
lt0Yf8Wh5O | Differentiable Robot Rendering | https://openreview.net/forum?id=lt0Yf8Wh5O | [
"Ruoshi Liu",
"Alper Canberk",
"Shuran Song",
"Carl Vondrick"
] | null | null | Vision foundation models trained on massive amounts of visual data have shown unprecedented reasoning and planning skills in open-world settings. A key challenge in applying them to robotic tasks is the modality gap between visual data and action data. We introduce differentiable robot rendering, a method allowing the ... | [
"Robot Representation",
"Visual Foundation Model"
] | We introduce a differentiable robot rendering method based on deformable Gaussians splattings and show many downstream applications. Abstract: | 72 | 2410.13851 | title_snapshot |
y8XkuQIrvI | MILES: Making Imitation Learning Easy with Self-Supervision | https://openreview.net/forum?id=y8XkuQIrvI | [
"Georgios Papagiannis",
"Edward Johns"
] | null | null | Data collection in imitation learning often requires significant, laborious human supervision, such as numerous demonstrations, and/or frequent environment resets for methods that incorporate reinforcement learning. In this work, we propose an alternative approach, MILES: a fully autonomous, self-supervised data collec... | [
"Imitation Learning",
"Robotic Manipulation",
"Self-Supervised Data Collection"
] | MILES is an imitation learning method that collects data in a self-supervised manner to train policies that can learn complex manipulation skills from a single demonstration. | 74 | 2410.19693 | title_snapshot |
5u9l6U61S7 | GenSim2: Scaling Robot Data Generation with Multi-modal and Reasoning LLMs | https://openreview.net/forum?id=5u9l6U61S7 | [
"Pu Hua",
"Minghuan Liu",
"Annabella Macaluso",
"Yunfeng Lin",
"Weinan Zhang",
"Huazhe Xu",
"Lirui Wang"
] | null | null | Robotic simulation today remains challenging to scale up due to the human efforts required to create diverse simulation tasks and scenes. Simulation-trained policies also face scalability issues as many sim-to-real methods focus on a single task. To address these challenges, this work proposes GenSim2, a scalable frame... | [
"Generative Simulation; Robotics; Learning"
] | We propose GenSim2, a scalable framework that uses foundation models for diverse, complex, realistic simulation tasks and data generation. | 75 | 2410.03645 | title_snapshot |
U5RPcnFhkq | FetchBench: A Simulation Benchmark for Robot Fetching | https://openreview.net/forum?id=U5RPcnFhkq | [
"Beining Han",
"Meenal Parakh",
"Derek Geng",
"Jack A Defay",
"Gan Luyang",
"Jia Deng"
] | null | null | Fetching, which includes approaching, grasping, and retrieving, is a critical challenge for robot manipulation tasks. Existing methods primarily focus on table-top scenarios, which do not adequately capture the complexities of environments where both grasping and planning are essential. To address this gap, we propose ... | [
"Grasping; Benchmark; Imitation Learning"
] | null | 79 | 2406.11793 | title_snapshot |
S8jQtafbT3 | Autonomous Interactive Correction MLLM for Robust Robotic Manipulation | https://openreview.net/forum?id=S8jQtafbT3 | [
"Chuyan Xiong",
"Chengyu Shen",
"Xiaoqi Li",
"Kaichen Zhou",
"Jiaming Liu",
"Ruiping Wang",
"Hao Dong"
] | null | null | The ability to reflect on and correct failures is crucial for robotic systems to interact stably with real-life objects. Observing the generalization and reasoning capabilities of Multimodal Large Language Models (MLLMs), previous approaches have aimed to utilize these models to enhance robotic systems accordingly. How... | [
"large language model",
"robotics"
] | null | 86 | 2406.11548 | title_judge |
CpXiqz6qf4 | SonicSense: Object Perception from In-Hand Acoustic Vibration | https://openreview.net/forum?id=CpXiqz6qf4 | [
"Jiaxun Liu",
"Boyuan Chen"
] | null | null | We introduce SonicSense, a holistic design of hardware and software to enable rich robot object perception through in-hand acoustic vibration sensing. While previous studies have shown promising results with acoustic sensing for object perception, current solutions are constrained to a handful of objects with simple ge... | [
"Tactile Perception",
"Object State Estimation",
"Audio",
"Acoustic Vibration Sensing"
] | SonicSense, an integrated hardware and software solution to enable rich object perception capabilities with in-hand acoustic vibration for a multi-finger robot hand. | 94 | 2406.17932 | title_snapshot |
fC0wWeXsVm | Learning Robot Soccer from Egocentric Vision with Deep Reinforcement Learning | https://openreview.net/forum?id=fC0wWeXsVm | [
"Dhruva Tirumala",
"Markus Wulfmeier",
"Ben Moran",
"Sandy Huang",
"Jan Humplik",
"Guy Lever",
"Tuomas Haarnoja",
"Leonard Hasenclever",
"Arunkumar Byravan",
"Nathan Batchelor",
"Neil sreendra",
"Kushal Patel",
"Marlon Gwira",
"Francesco Nori",
"Martin Riedmiller",
"Nicolas Heess"
] | null | null | We apply multi-agent deep reinforcement learning (RL) to train end-to-end robot soccer policies with fully onboard computation and sensing via egocentric RGB vision. This setting reflects many challenges of real-world robotics, including active perception, agile full-body control, and long-horizon planning in a dynamic... | [
"robotics",
"deep reinforcement learning"
] | null | 95 | 2405.02425 | title_snapshot |
AzP6kSEffm | Dynamics-Guided Diffusion Model for Sensor-less Robot Manipulator Design | https://openreview.net/forum?id=AzP6kSEffm | [
"Xiaomeng Xu",
"Huy Ha",
"Shuran Song"
] | null | null | We present Dynamics-Guided Diffusion Model (DGDM), a data-driven framework for generating task-specific manipulator designs without task-specific training. Given object shapes and task specifications, DGDM generates sensor-less manipulator designs that can blindly manipulate objects towards desired motions and poses u... | [
"manipulator design",
"hardware optimization",
"diffusion model"
] | Generating task-specific manipulator designs without task-specific training | 98 | 2402.15038 | title_snapshot |
ctzBccpolr | RoVi-Aug: Robot and Viewpoint Augmentation for Cross-Embodiment Robot Learning | https://openreview.net/forum?id=ctzBccpolr | [
"Lawrence Yunliang Chen",
"Chenfeng Xu",
"Karthik Dharmarajan",
"Richard Cheng",
"Kurt Keutzer",
"Masayoshi Tomizuka",
"Quan Vuong",
"Ken Goldberg"
] | null | null | Scaling up robot learning requires large and diverse datasets, and how to efficiently reuse collected data and transfer policies to new embodiments remains an open question. Emerging research such as the Open-X Embodiment (OXE) project has shown promise in leveraging skills by combining datasets including different rob... | [
"Cross-Embodiment Learning",
"Viewpoint Robust",
"Data Augmentation"
] | data augmentation that uses diffusion models to generate novel robots and viewpoints | 100 | 2409.03403 | title_snapshot |
7c5rAY8oU3 | Automated Creation of Digital Cousins for Robust Policy Learning | https://openreview.net/forum?id=7c5rAY8oU3 | [
"Tianyuan Dai",
"Josiah Wong",
"Yunfan Jiang",
"Chen Wang",
"Cem Gokmen",
"Ruohan Zhang",
"Jiajun Wu",
"Li Fei-Fei"
] | null | null | Training robot policies in the real world can be unsafe, costly, and difficult to scale. Simulation serves as an inexpensive and potentially limitless source of training data, but suffers from the semantics and physics disparity between simulated and real-world environments. These discrepancies can be minimized by trai... | [
"Real-to-Sim; Digital Twin; Sim-to-Real Transfer"
] | An end-to-end framework for automatically generating fully interactive simulated scenes ("digital cousins") from a single real-world RGB image enabling training of robot policies that can be deployed zero-shot in the real world. | 104 | 2410.07408 | title_snapshot |
eeoX7tCoK2 | Shelf-Supervised Cross-Modal Pre-Training for 3D Object Detection | https://openreview.net/forum?id=eeoX7tCoK2 | [
"Mehar Khurana",
"Neehar Peri",
"James Hays",
"Deva Ramanan"
] | null | null | State-of-the-art 3D object detectors are often trained on massive labeled datasets. However, annotating 3D bounding boxes remains prohibitively expensive and time-consuming, particularly for LiDAR. Instead, recent works demonstrate that self-supervised pre-training with unlabeled data can improve detection accuracy wit... | [
"Shelf-Supervised 3D Object Detection",
"Vision-Language Models",
"Autonomous Vehicles"
] | We demonstrate that pre-training detectors with zero-shot 3D bounding box pseudo-labels achieves higher semi-supervised detection accuracy than prior contrastive learning objectives. | 113 | 2406.10115 | title_snapshot |
VUhlMfEekm | Implicit Grasp Diffusion: Bridging the Gap between Dense Prediction and Sampling-based Grasping | https://openreview.net/forum?id=VUhlMfEekm | [
"Pinhao Song",
"Pengteng Li",
"Renaud Detry"
] | null | null | There are two dominant approaches in modern robot grasp planning: dense prediction and sampling-based methods. Dense prediction calculates viable grasps across the robot’s view but is limited to predicting one grasp per voxel. Sampling-based methods, on the other hand, encode multi-modal grasp distributions, allowing f... | [
"Grasping",
"Implicit Neural Representations",
"Diffusion Models"
] | Using diffusion models to generate grasps based on local features | 118 | null | null |
eJHy0AF5TO | RiEMann: Near Real-Time SE(3)-Equivariant Robot Manipulation without Point Cloud Segmentation | https://openreview.net/forum?id=eJHy0AF5TO | [
"Chongkai Gao",
"Zhengrong Xue",
"Shuying Deng",
"Tianhai Liang",
"Siqi Yang",
"Lin Shao",
"Huazhe Xu"
] | null | null | We present RiEMann, an end-to-end near Real-time SE(3)-Equivariant Robot Manipulation imitation learning framework from scene point cloud input. Compared to previous methods that rely on descriptor field matching, RiEMann directly predicts the target actions for manipulation without any object segmentation. RiEMann can... | [
"SE(3)-Equivariance",
"Manipulation",
"Imitation Learning"
] | This paper proposes RiEMann, an efficient and near real-time SE(3)-equivariant robot imitation learning framework for manipulation tasks framework without point cloud segmentation. | 119 | 2403.19460 | title_snapshot |
EyEE7547vy | Event3DGS: Event-Based 3D Gaussian Splatting for High-Speed Robot Egomotion | https://openreview.net/forum?id=EyEE7547vy | [
"Tianyi Xiong",
"Jiayi Wu",
"Botao He",
"Cornelia Fermuller",
"Yiannis Aloimonos",
"Heng Huang",
"Christopher Metzler"
] | null | null | By combining differentiable rendering with explicit point-based scene representations, 3D Gaussian Splatting (3DGS) has demonstrated breakthrough 3D reconstruction capabilities.
However, to date 3DGS has had limited impact on robotics, where high-speed egomotion is pervasive: Egomotion introduces motion blur and l... | [
"Event-based 3D Reconstruction",
"Gaussian Splatting",
"High-speed Robot Egomotion"
] | null | 126 | 2406.02972 | title_snapshot |
cocHfT7CEs | Generative Image as Action Models | https://openreview.net/forum?id=cocHfT7CEs | [
"Mohit Shridhar",
"Yat Long Lo",
"Stephen James"
] | null | null | Image-generation diffusion models have been fine-tuned to unlock new capabilities such as image-editing and novel view synthesis. Can we similarly unlock image-generation models for visuomotor control? We present GENIMA, a behavior-cloning agent that fine-tunes Stable Diffusion to “draw joint-actions” as targets on RGB... | [
"Diffusion Models",
"Image Generation",
"Behavior Cloning",
"Visuomotor"
] | Stable Diffusion can be fine-tuned to draw joint-actions for visuomotor control. | 132 | 2407.07875 | title_snapshot |
WjDR48cL3O | Continuous Control with Coarse-to-fine Reinforcement Learning | https://openreview.net/forum?id=WjDR48cL3O | [
"Younggyo Seo",
"Jafar Uruç",
"Stephen James"
] | null | null | Despite recent advances in improving the sample-efficiency of reinforcement learning (RL) algorithms, designing an RL algorithm that can be practically deployed in real-world environments remains a challenge. In this paper, we present Coarse-to-fine Reinforcement Learning (CRL), a framework that trains RL agents to zoo... | [
"Reinforcement Learning",
"Sample-Efficient",
"Action Discretization"
] | We present a sample-efficient RL algorithm that can be deployed in real-robot experiments. We train RL agents to zoom-into a continuous action space in a coarse-to-fine manner. | 133 | 2407.07787 | title_snapshot |
2LLu3gavF1 | Robot See Robot Do: Imitating Articulated Object Manipulation with Monocular 4D Reconstruction | https://openreview.net/forum?id=2LLu3gavF1 | [
"Justin Kerr",
"Chung Min Kim",
"Mingxuan Wu",
"Brent Yi",
"Qianqian Wang",
"Ken Goldberg",
"Angjoo Kanazawa"
] | null | null | Humans can learn to manipulate new objects by simply watching others; providing robots with the ability to learn from such demonstrations would enable a natural interface specifying new behaviors. This work develops Robot See Robot Do (RSRD), a method for imitating articulated object manipulation from a single monocula... | [
"Feature Fields",
"Visual Imitation",
"Grasping",
"Articulated Objects"
] | We show that part-centric feature fields can be used for articulated object manipulation from a single human video. | 136 | 2409.18121 | title_snapshot |
GVX6jpZOhU | RoboPoint: A Vision-Language Model for Spatial Affordance Prediction in Robotics | https://openreview.net/forum?id=GVX6jpZOhU | [
"Wentao Yuan",
"Jiafei Duan",
"Valts Blukis",
"Wilbert Pumacay",
"Ranjay Krishna",
"Adithyavairavan Murali",
"Arsalan Mousavian",
"Dieter Fox"
] | null | null | From rearranging objects on a table to putting groceries into shelves, robots must plan precise action points to perform tasks accurately and reliably. In spite of the recent adoption of vision language models (VLMs) to control robot behavior, VLMs struggle to precisely articulate robot actions using language. We intro... | [
"Foundation Model",
"Affordance Prediction",
"Open-world Manipulation"
] | A Vision-Language Model that predicts image-based keypoint affordance conditioned on language instructions | 138 | 2406.10721 | title_judge |
3wBqoPfoeJ | Twisting Lids Off with Two Hands | https://openreview.net/forum?id=3wBqoPfoeJ | [
"Toru Lin",
"Zhao-Heng Yin",
"Haozhi Qi",
"Pieter Abbeel",
"Jitendra Malik"
] | null | null | Manipulating objects with two multi-fingered hands has been a long-standing challenge in robotics, due to the contact-rich nature of many manipulation tasks and the complexity inherent in coordinating a high-dimensional bimanual system. In this work, we share novel insights into physical modeling, real-time perception,... | [
"Bimanual Manipulation",
"Sim-to-Real",
"Reinforcement Learning"
] | null | 142 | 2403.02338 | title_snapshot |
dsxmR6lYlg | Reinforcement Learning with Foundation Priors: Let Embodied Agent Efficiently Learn on Its Own | https://openreview.net/forum?id=dsxmR6lYlg | [
"Weirui Ye",
"Yunsheng Zhang",
"Haoyang Weng",
"Xianfan Gu",
"Shengjie Wang",
"Tong Zhang",
"Mengchen Wang",
"Pieter Abbeel",
"Yang Gao"
] | null | null | Reinforcement learning (RL) is a promising approach for solving robotic manipulation tasks.
However, it is challenging to apply the RL algorithms directly in the real world.
For one thing, RL is data-intensive and typically requires millions of interactions with environments, which are impractical in real scenarios.
F... | [
"Reinforcement Learning",
"Foundation Models",
"Robotics",
"VLMs"
] | null | 144 | 2310.02635 | title_judge |
5iXG6EgByK | Promptable Closed-loop Traffic Simulation | https://openreview.net/forum?id=5iXG6EgByK | [
"Shuhan Tan",
"Boris Ivanovic",
"Yuxiao Chen",
"Boyi Li",
"Xinshuo Weng",
"Yulong Cao",
"Philipp Kraehenbuehl",
"Marco Pavone"
] | null | null | Simulation stands as a cornerstone for safe and efficient autonomous driving development. At its core a simulation system ought to produce realistic, reactive, and controllable traffic patterns. In this paper, we propose ProSim, a multimodal promptable closed-loop traffic simulation framework. ProSim allows the user to... | [
"Autonomous Driving",
"Scenario Generation",
"Traffic Simulation"
] | 1) We propose ProSim, a multimodal promptable closed-loop traffic simulation framework. 2) We create ProSim-Instruct-520k, a large-scale multimodal prompt-scenario paired driving dataset. | 145 | 2409.05863 | title_snapshot |
O05tIQt2d5 | TOP-Nav: Legged Navigation Integrating Terrain, Obstacle and Proprioception Estimation | https://openreview.net/forum?id=O05tIQt2d5 | [
"Junli Ren",
"Yikai Liu",
"Yingru Dai",
"Junfeng Long",
"Guijin Wang"
] | null | null | Legged navigation is typically examined within open-world, off-road, and challenging environments. In these scenarios, estimating external disturbances requires a complex synthesis of multi-modal information. This underlines a major limitation in existing works that primarily focus on avoiding obstacles. In this work, ... | [
"Navigation",
"Task Planning",
"Reinforcement Learning"
] | We present a legged navigation system that achieves open-world navigation surpassing limitations posed by visual conditions or prior knowledge by integrating terrain, obstacle and proprioception. | 147 | 2404.15256 | title_snapshot |
cvAIaS6V2I | OPEN TEACH: A Versatile Teleoperation System for Robotic Manipulation | https://openreview.net/forum?id=cvAIaS6V2I | [
"Aadhithya Iyer",
"Zhuoran Peng",
"Yinlong Dai",
"Irmak Guzey",
"Siddhant Haldar",
"Soumith Chintala",
"Lerrel Pinto"
] | null | null | Open-sourced, user-friendly tools form the bedrock of scientific advancement across disciplines. The widespread adoption of data-driven learning has led to remarkable progress in multi-fingered dexterity, bimanual manipulation, and applications ranging from logistics to home robotics. However, existing data collection ... | [
"Teleoperation",
"Robot Learning",
"Robotic Manipulation"
] | Versatile Teleoperation System for Robotic Manipulation | 149 | 2403.07870 | title_snapshot |
uEbJXWobif | EXTRACT: Efficient Policy Learning by Extracting Transferable Robot Skills from Offline Data | https://openreview.net/forum?id=uEbJXWobif | [
"Jesse Zhang",
"Minho Heo",
"Zuxin Liu",
"Erdem Biyik",
"Joseph J Lim",
"Yao Liu",
"Rasool Fakoor"
] | null | null | Most reinforcement learning (RL) methods focus on learning optimal policies over low-level action spaces. While these methods can perform well in their training environments, they lack the flexibility to transfer to new tasks.
Instead, RL agents that can act over useful, temporally extended skills rather than low-leve... | [
"reinforcement learning",
"skill-based reinformement learning",
"skill learning",
"transfer learning",
"foundation models for robotics",
"robot learning"
] | We extract discrete, meaningfully-aligned skills from offline data for efficient reinforcement learning of new tasks. | 152 | 2406.17768 | title_snapshot |
Qz2N4lWBk3 | Learning Granular Media Avalanche Behavior for Indirectly Manipulating Obstacles on a Granular Slope | https://openreview.net/forum?id=Qz2N4lWBk3 | [
"Haodi Hu",
"Feifei Qian",
"Daniel Seita"
] | null | null | Legged robot locomotion on sand slopes is challenging due to the complex dynamics of granular media and how the lack of solid surfaces can hinder locomotion. A promising strategy, inspired by ghost crabs and other organisms in nature, is to strategically interact with rocks, debris, and other obstacles to facilitate mo... | [
"Granular media",
"Avalanche dynamics",
"Legged robots."
] | In this work, we propose Granular Robotic Avalanche INteraction (GRAIN), a novel learning-based method for leveraging granular avalanche dynamics for indirectly manipulating objects on a granular slope. | 153 | 2407.01898 | title_snapshot |
KXsropnmNI | Transferable Tactile Transformers for Representation Learning Across Diverse Sensors and Tasks | https://openreview.net/forum?id=KXsropnmNI | [
"Jialiang Zhao",
"Yuxiang Ma",
"Lirui Wang",
"Edward Adelson"
] | null | null | This paper presents T3: Transferable Tactile Transformers, a framework for tactile representation learning that scales across multi-sensors and multi-tasks.T3 is designed to overcome the contemporary issue that camera-based tactile sensing is extremely heterogeneous, i.e. sensors are built into different form factors, ... | [
"Tactile Sensing",
"Representation Learning",
"Heterogeneous Learning",
"Robot Manipulation",
"Robot Learning"
] | A framework that learns a tactile representation that transfers between diverse sensors and tasks, and an aggregated tactile dataset that is the largest and most diverse to date. | 155 | 2406.13640 | title_snapshot |
N5IS6DzBmL | Play to the Score: Stage-Guided Dynamic Multi-Sensory Fusion for Robotic Manipulation | https://openreview.net/forum?id=N5IS6DzBmL | [
"Ruoxuan Feng",
"Di Hu",
"Wenke Ma",
"Xuelong Li"
] | null | null | Humans possess a remarkable talent for flexibly alternating to different senses when interacting with the environment. Picture a chef skillfully gauging the timing of ingredient additions and controlling the heat according to the colors, sounds, and aromas, seamlessly navigating through every stage of the complex cooki... | [
"Multi-Sensory",
"Robotic Manipulation",
"Multi-Stage"
] | We propose MS-Bot, a stage-guided dynamic multi-sensory fusion method with coarse-to-fine stage understanding, which dynamically adjusts the priority of modalities based on the fine-grained state within the predicted current stage. | 156 | 2408.01366 | title_snapshot |
SFJz5iLvur | Lessons from Learning to Spin “Pens” | https://openreview.net/forum?id=SFJz5iLvur | [
"Jun Wang",
"Ying Yuan",
"Haichuan Che",
"Haozhi Qi",
"Yi Ma",
"Jitendra Malik",
"Xiaolong Wang"
] | null | null | In-hand manipulation of pen-like objects is a most basic and important skill in our daily lives, as many tools such as hammers and screwdrivers are similarly shaped. However, current learning-based methods struggle with this task due to a lack of high-quality demonstrations and the significant gap between simulation an... | [
"Dexterous In-Hand Manipulation",
"Reinforcement Learning"
] | null | 158 | 2407.18902 | title_snapshot |
VFs1vbQnYN | Sim-to-Real Transfer via 3D Feature Fields for Vision-and-Language Navigation | https://openreview.net/forum?id=VFs1vbQnYN | [
"Zihan Wang",
"Xiangyang Li",
"Jiahao Yang",
"Yeqi Liu",
"Shuqiang Jiang"
] | null | null | Vision-and-language navigation (VLN) enables the agent to navigate to a remote location in 3D environments following the natural language instruction. In this field, the agent is usually trained and evaluated in the navigation simulators, lacking effective approaches for sim-to-real transfer. The VLN agents with only a... | [
"Vision-and-Language Navigation",
"3D Feature Fields",
"Semantic Traversable Map"
] | null | 160 | 2406.09798 | title_snapshot |
LmOF7UAOZ7 | A Planar-Symmetric SO(3) Representation for Learning Grasp Detection | https://openreview.net/forum?id=LmOF7UAOZ7 | [
"Tianyi Ko",
"Takuya Ikeda",
"Hiroya Sato",
"Koichi Nishiwaki"
] | null | null | Planar-symmetric hands, such as parallel grippers, are widely adopted in both research and industrial fields.
Their symmetry, however, introduces ambiguity and discontinuity in the SO(3) representation, which hinders both the training and inference of neural network-based grasp detectors.
We propose a novel SO(3) repre... | [
"Grasp Detection",
"Rotation Representation",
"Parallel Gripper"
] | We propose a novel SO(3) representation that can parametrize a pair of planar-symmetric poses with a single parameter set by leveraging the 2D Bingham distribution. | 162 | 2410.04826 | title_snapshot |
xYleTh2QhS | Adaptive Diffusion Terrain Generator for Autonomous Uneven Terrain Navigation | https://openreview.net/forum?id=xYleTh2QhS | [
"Youwei Yu",
"Junhong Xu",
"Lantao Liu"
] | null | null | Model-free reinforcement learning has emerged as a powerful method for developing robust robot control policies capable of navigating through complex and unstructured terrains.
The effectiveness of these methods hinges on two essential elements:
(1) the use of massively parallel physics simulations to expedite policy ... | [
"Curriculum Reinforcement Learning",
"Diffusion Model",
"Field Robotics"
] | Our work introduces Adaptive Diffusion Terrain Generator, leveraging DDPMs to dynamically generate complex, realistic terrains for improved policy training and generalization to real-world environments. | 166 | 2410.10766 | title_snapshot |
bk28WlkqZn | 3D-ViTac: Learning Fine-Grained Manipulation with Visuo-Tactile Sensing | https://openreview.net/forum?id=bk28WlkqZn | [
"Binghao Huang",
"Yixuan Wang",
"Xinyi Yang",
"Yiyue Luo",
"Yunzhu Li"
] | null | null | Tactile and visual perception are both crucial for humans to perform fine-grained interactions with their environment. Developing similar multi-modal sensing capabilities for robots can significantly enhance and expand their manipulation skills. This paper introduces **3D-ViTac**, a multi-modal sensing and learning sys... | [
"Contact-Rich Manipulation",
"Multi-Modal Perception",
"Tactile Sensing",
"Imitation Learning"
] | null | 170 | 2410.24091 | title_snapshot |
ylZHvlwUcI | Theia: Distilling Diverse Vision Foundation Models for Robot Learning | https://openreview.net/forum?id=ylZHvlwUcI | [
"Jinghuan Shang",
"Karl Schmeckpeper",
"Brandon B. May",
"Maria Vittoria Minniti",
"Tarik Kelestemur",
"David Watkins",
"Laura Herlant"
] | null | null | Vision-based robot policy learning, which maps visual inputs to actions, necessitates a holistic understanding of diverse visual tasks beyond single-task needs like classification or segmentation. Inspired by this, we introduce Theia, a vision foundation model for robot learning that distills multiple off-the-shelf vis... | [
"visual representation",
"robot learning",
"distillation",
"foundation model"
] | We build a robot vision foundation model by distilling existing vision foundation models, which improves downstream robot learning performance. | 171 | 2407.20179 | title_snapshot |
8XFT1PatHy | Splat-MOVER: Multi-Stage, Open-Vocabulary Robotic Manipulation via Editable Gaussian Splatting | https://openreview.net/forum?id=8XFT1PatHy | [
"Olaolu Shorinwa",
"Johnathan Tucker",
"Aliyah Smith",
"Aiden Swann",
"Timothy Chen",
"Roya Firoozi",
"Monroe David Kennedy",
"Mac Schwager"
] | null | null | We present Splat-MOVER, a modular robotics stack for open-vocabulary
robotic manipulation, which leverages the editability of Gaussian Splatting (GSplat)
scene representations to enable multi-stage manipulation tasks. Splat-MOVER
consists of: (i) ASK-Splat, a GSplat representation that distills semantic and grasp
affor... | [
"Gaussian Splatting",
"Robotic Grasping",
"Robotic Manipulation",
"Scene Editing"
] | We present Splat-MOVER, a modular robotics stack for open-vocabulary robotic manipulation, which leverages the editability of Gaussian Splatting scene representations to enable multi-stage manipulation tasks. | 173 | 2405.04378 | title_snapshot |
KULBk5q24a | CoViS-Net: A Cooperative Visual Spatial Foundation Model for Multi-Robot Applications | https://openreview.net/forum?id=KULBk5q24a | [
"Jan Blumenkamp",
"Steven Morad",
"Jennifer Gielis",
"Amanda Prorok"
] | null | null | Autonomous robot operation in unstructured environments is often underpinned by spatial understanding through vision. Systems composed of multiple concurrently operating robots additionally require access to frequent, accurate and reliable pose estimates. Classical vision-based methods to regress relative pose are comm... | [
"Multi-Robot Systems",
"Robot Perception",
"Foundation Models"
] | We propose a decentralized, platform-agnostic visual spatial foundation model for multi-robot systems, enhancing spatial understanding through data-derived priors for real-world, online deployment. | 176 | 2405.01107 | title_snapshot |
nmEt0ci8hi | General Flow as Foundation Affordance for Scalable Robot Learning | https://openreview.net/forum?id=nmEt0ci8hi | [
"Chengbo Yuan",
"Chuan Wen",
"Tong Zhang",
"Yang Gao"
] | null | null | We address the challenge of acquiring real-world manipulation skills with a scalable framework. We hold the belief that identifying an appropriate prediction target capable of leveraging large-scale datasets is crucial for achieving efficient and universal learning.
Therefore, we propose to utilize 3D flow, which repre... | [
"Flow",
"Transferable Affordance",
"Scalability"
] | A flow-based framework enables real-world zero-shot human-to-robot skill transfer while providing scalability, universality, and stability for robot learning. | 178 | 2401.11439 | title_snapshot |
rvKWXxIvj0 | Non-rigid Relative Placement through 3D Dense Diffusion | https://openreview.net/forum?id=rvKWXxIvj0 | [
"Eric Cai",
"Octavian Donca",
"Ben Eisner",
"David Held"
] | null | null | The task of "relative placement" is to predict the placement of one object in relation to another, e.g. placing a mug on a mug rack. Recent methods for relative placement have made tremendous progress towards data-efficient learning for robot manipulation; using explicit object-centric geometric reasoning, these approa... | [
"Deformable",
"Non-rigid",
"Manipulation",
"Relative Placement"
] | We expand the "relative placement" formulation beyond rigid body manipulation, allowing for generalization to novel object instances, unseen scene configurations, and multimodal placements for highly deformable tasks. | 180 | 2410.19247 | title_snapshot |
3bcujpPikC | FREA: Feasibility-Guided Generation of Safety-Critical Scenarios with Reasonable Adversariality | https://openreview.net/forum?id=3bcujpPikC | [
"Keyu Chen",
"Yuheng Lei",
"Hao Cheng",
"Haoran Wu",
"Wenchao Sun",
"Sifa Zheng"
] | null | null | Generating safety-critical scenarios, which are essential yet difficult to collect at scale, offers an effective method to evaluate the robustness of autonomous vehicles (AVs). Existing methods focus on optimizing adversariality while preserving the naturalness of scenarios, aiming to achieve a balance through data-dri... | [
"Feasibility",
"Scenario Generation",
"Autonomous Driving"
] | FREA incorporates feasibility as guidance to generate adversarial yet AV-feasible, safety-critical scenarios for autonomous driving. | 182 | 2406.02983 | title_snapshot |
wcbrhPnOei | RobotKeyframing: Learning Locomotion with High-Level Objectives via Mixture of Dense and Sparse Rewards | https://openreview.net/forum?id=wcbrhPnOei | [
"Fatemeh Zargarbashi",
"Jin Cheng",
"Dongho Kang",
"Robert Sumner",
"Stelian Coros"
] | null | null | This paper presents a novel learning-based control framework that uses keyframing to incorporate high-level objectives in natural locomotion for legged robots. These high-level objectives are specified as a variable number of partial or complete pose targets that are spaced arbitrarily in time. Our proposed framework u... | [
"Legged robots",
"Multi-Critic Reinforcement Learning",
"Motion Imitation"
] | This paper presents a novel learning-based control framework that incorporates high-level objectives for the natural locomotion of legged robots through keyframing. | 183 | 2407.11562 | title_snapshot |
0gDbaEtVrd | One Model to Drift Them All: Physics-Informed Conditional Diffusion Model for Driving at the Limits | https://openreview.net/forum?id=0gDbaEtVrd | [
"Franck Djeumou",
"Thomas Jonathan Lew",
"NAN DING",
"Michael Thompson",
"Makoto Suminaka",
"Marcus Greiff",
"John Subosits"
] | null | null | Enabling autonomous vehicles to reliably operate at the limits of handling— where tire forces are saturated — would improve their safety, particularly in scenarios like emergency obstacle avoidance or adverse weather conditions.
However, unlocking this capability is challenging due to the task's dynamic nature and the ... | [
"Diffusion Models",
"Learning for Control",
"Autonomous Drifting",
"Model Predictive Control"
] | A single diffusion model can adapt online to reliably control a Toyota Supra and a Lexus LC 500 on challenging autonomous drifting tasks | 184 | null | null |
Si2krRESZb | TieBot: Learning to Knot a Tie from Visual Demonstration through a Real-to-Sim-to-Real Approach | https://openreview.net/forum?id=Si2krRESZb | [
"Weikun Peng",
"Jun Lv",
"Yuwei Zeng",
"Haonan Chen",
"Siheng Zhao",
"Jichen Sun",
"Cewu Lu",
"Lin Shao"
] | null | null | The tie-knotting task is highly challenging due to the tie's high deformation and long-horizon manipulation actions. This work presents TieBot, a Real-to-Sim-to-Real learning from visual demonstration system for the robots to learn to knot a tie. We introduce the Hierarchical Feature Matching approach to estimate a seq... | [
"cloth manipulation",
"learning from demonstration",
"robot learning"
] | null | 186 | 2407.03245 | title_snapshot |
Czs2xH9114 | WoCoCo: Learning Whole-Body Humanoid Control with Sequential Contacts | https://openreview.net/forum?id=Czs2xH9114 | [
"Chong Zhang",
"Wenli Xiao",
"Tairan He",
"Guanya Shi"
] | null | null | Humanoid activities involving sequential contacts are crucial for complex robotic interactions and operations in the real world and are traditionally solved by model-based motion planning, which is time-consuming and often relies on simplified dynamics models.
Although model-free reinforcement learning (RL) has become... | [
"Whole-Body Humanoid Control",
"Multi-Contact Control",
"Reinforcement Learning"
] | We propose a unified RL framework to learn whole-body humanoid control with sequential contacts. | 187 | 2406.06005 | title_snapshot |
ubq7Co6Cbv | Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networks | https://openreview.net/forum?id=ubq7Co6Cbv | [
"Alex Quach",
"Makram Chahine",
"Alexander Amini",
"Ramin Hasani",
"Daniela Rus"
] | null | null | Simulators are powerful tools for autonomous robot learning as they offer scalable data generation, flexible design, and optimization of trajectories.
However, transferring behavior learned from simulation data into the real world proves to be difficult, usually mitigated with compute-heavy domain randomization method... | [
"End-to-end learning",
"Gaussian Splatting",
"Sim-to-real transfer"
] | This work enhances sim-to-real quadrotor navigation by combining Gaussian Splatting with Liquid neural networks, achieving robust real-world performance and adaptability to environmental changes. | 188 | 2406.15149 | title_snapshot |
UHxPZgK33I | RoboEXP: Action-Conditioned Scene Graph via Interactive Exploration for Robotic Manipulation | https://openreview.net/forum?id=UHxPZgK33I | [
"Hanxiao Jiang",
"Binghao Huang",
"Ruihai Wu",
"Zhuoran Li",
"Shubham Garg",
"Hooshang Nayyeri",
"Shenlong Wang",
"Yunzhu Li"
] | null | null | We introduce the novel task of interactive scene exploration, wherein robots autonomously explore environments and produce an action-conditioned scene graph (ACSG) that captures the structure of the underlying environment. The ACSG accounts for both low-level information (geometry and semantics) and high-level informat... | [
"Action-Conditioned Scene Graph",
"Foundation Models for Robotics",
"Scene Exploration",
"Robotic Manipulation"
] | null | 192 | 2402.15487 | title_snapshot |
fR1rCXjCQX | Learning Compositional Behaviors from Demonstration and Language | https://openreview.net/forum?id=fR1rCXjCQX | [
"Weiyu Liu",
"Neil Nie",
"Ruohan Zhang",
"Jiayuan Mao",
"Jiajun Wu"
] | null | null | We introduce Behavior from Language and Demonstration (BLADE), a framework for long-horizon robotic manipulation by integrating imitation learning and model-based planning. BLADE leverages language-annotated demonstrations, extracts abstract action knowledge from large language models (LLMs), and constructs a library o... | [
"Manipulation",
"Planning Abstractions",
"Learning from Language"
] | We introduce a new framework for long-horizon robotic manipulation by integrating imitation learning and planning with models automatically constructed from language. | 195 | 2505.21981 | title_snapshot |
EiqQEsOMZt | TaMMa: Target-driven Multi-subscene Mobile Manipulation | https://openreview.net/forum?id=EiqQEsOMZt | [
"Jiawei Hou",
"Tianyu Wang",
"Tongying Pan",
"Shouyan Wang",
"Xiangyang Xue",
"Yanwei Fu"
] | null | null | For everyday service robotics, the ability to navigate back and forth based on tasks in multi-subscene environments and perform delicate manipulations is crucial and highly practical.
While existing robotics primarily focus on complex tasks within a single scene or simple tasks across scalable scenes individually, robo... | [
"Multi-subscene",
"3D Gaussians",
"Scene Inpainting",
"Target-driven Mobile Manipulation"
] | null | 196 | null | null |
AEq0onGrN2 | Physically Embodied Gaussian Splatting: A Visually Learnt and Physically Grounded 3D Representation for Robotics | https://openreview.net/forum?id=AEq0onGrN2 | [
"Jad Abou-Chakra",
"Krishan Rana",
"Feras Dayoub",
"Niko Suenderhauf"
] | null | null | For robots to robustly understand and interact with the physical world, it is highly beneficial to have a comprehensive representation -- modelling geometry, physics, and visual observations -- that informs perception, planning, and control algorithms. We propose a novel dual "Gaussian-Particle" representation that mo... | [
"3D Representation",
"Gaussian Splatting",
"Robotics",
"Tracking",
"Physics"
] | We use particles to represent the physical state of a robot's world and 3D Gaussians for its visual state, combining them to enable forward modeling and realtime visual-based correction from images coming from 3 cameras. | 199 | 2406.10788 | title_judge |
uHdVI3QMr6 | A Dual Approach to Imitation Learning from Observations with Offline Datasets | https://openreview.net/forum?id=uHdVI3QMr6 | [
"Harshit Sikchi",
"Caleb Chuck",
"Amy Zhang",
"Scott Niekum"
] | null | null | Demonstrations are an effective alternative to task specification for learning agents in settings where designing a reward function is difficult. However, demonstrating expert behavior in the action space of the agent becomes unwieldy when robots have complex, unintuitive morphologies. We consider the practical setting... | [
"Learning from Observations",
"Imitation Learning"
] | A dual approach to LfO that is principled, computationally efficient, and empirically performant. | 200 | 2406.08805 | title_snapshot |
9iG3SEbMnL | ReKep: Spatio-Temporal Reasoning of Relational Keypoint Constraints for Robotic Manipulation | https://openreview.net/forum?id=9iG3SEbMnL | [
"Wenlong Huang",
"Chen Wang",
"Yunzhu Li",
"Ruohan Zhang",
"Li Fei-Fei"
] | null | null | Representing robotic manipulation tasks as constraints that associate the robot and the environment is a promising way to encode desired robot behaviors. However, it remains unclear how to formulate the constraints such that they are 1) versatile to diverse tasks, 2) free of manual labeling, and 3) optimizable by off-t... | [
"Structural Representation",
"Model-Based Planning",
"Foundation Models"
] | Large vision models and vision-language models can generate a spatio-temporal series of constraints that map keypoints to a numerical cost, which are optimized to obtain actions for multi-stage, in-the-wild, bimanual, and reactive manipulation tasks. | 202 | 2409.01652 | title_snapshot |
qoebyrnF36 | Control with Patterns: A D-learning Method | https://openreview.net/forum?id=qoebyrnF36 | [
"Quan Quan",
"Kai-Yuan Cai",
"Chenyu Wang"
] | null | null | Learning-based control policies are widely used in various tasks in the field of robotics and control. However, formal (Lyapunov) stability guarantees for learning-based controllers with nonlinear dynamical systems are challenging to obtain.
We propose a novel control approach, namely Control with Patterns (CWP), to a... | [
"Lyapunov Methods",
"Reinforcement Learning",
"Control with Patterns",
"D-learning",
"Visual Servoing"
] | The proposed control with patterns based on D-learning is a type of Lyapunov function learning methods without requring any knowledge of the system dynamics, which parallels to Q-learning in reinforcement learning. | 203 | 2206.03809 | title_snapshot |
QtCtY8zl2T | Task Success Prediction for Open-Vocabulary Manipulation Based on Multi-Level Aligned Representations | https://openreview.net/forum?id=QtCtY8zl2T | [
"Miyu Goko",
"Motonari Kambara",
"Daichi Saito",
"Seitaro Otsuki",
"Komei Sugiura"
] | null | null | In this study, we consider the problem of predicting task success for open-vocabulary manipulation by a manipulator, based on instruction sentences and egocentric images before and after manipulation. Conventional approaches, including multimodal large language models (MLLMs), often fail to appropriately understand det... | [
"Task Success Prediction",
"Open-Vocabulary Manipulation",
"Multi-Level Aligned Visual Representation"
] | We introduce a task success prediction model for open-vocabulary manipulation. The model focuses on the differences between multi-level aligned representations of images. It outperformed existing models, including representative multimodal LLMs. | 210 | 2410.00436 | title_snapshot |
yYujuPxjDK | Language-guided Manipulator Motion Planning with Bounded Task Space | https://openreview.net/forum?id=yYujuPxjDK | [
"Thies Oelerich",
"Christian Hartl-Nesic",
"Andreas Kugi"
] | null | null | Language-based robot control is a powerful and versatile method to control a robot manipulator where large language models (LLMs) are used to reason about the environment. However, the generated robot motions by these controllers often lack safety and performance, resulting in jerky movements. In this work, a novel mod... | [
"Vision Language Models",
"Manipulation Planning",
"Path-following MPC"
] | null | 211 | null | null |
V5x0m6XDSV | Differentiable Discrete Elastic Rods for Real-Time Modeling of Deformable Linear Objects | https://openreview.net/forum?id=V5x0m6XDSV | [
"Yizhou Chen",
"Yiting Zhang",
"Zachary Brei",
"Tiancheng Zhang",
"Yuzhen Chen",
"Julie Wu",
"Ram Vasudevan"
] | null | null | This paper addresses the task of modeling Deformable Linear Objects (DLOs), such as ropes and cables, during dynamic motion over long time horizons. This task presents significant challenges due to the complex dynamics of DLOs. To address these challenges, this paper proposes differentiable Discrete Elastic Rods For de... | [
"Deformable Linear Objects Modeling",
"Physics-Informed Learning",
"Differentiable Simulation"
] | This paper proposes a novel framework that combines a differentiable physics-based model with learning framework to model DLOs accurately and in real-time. | 212 | 2406.05931 | title_snapshot |
1IzW0aniyg | EscIRL: Evolving Self-Contrastive IRL for Trajectory Prediction in Autonomous Driving | https://openreview.net/forum?id=1IzW0aniyg | [
"Siyue Wang",
"Zhaorun Chen",
"Zhuokai Zhao",
"Chaoli Mao",
"Yiyang Zhou",
"Jiayu He",
"Albert Sibo Hu"
] | null | null | While deep neural networks (DNN) and inverse reinforcement learning (IRL) have both been commonly used in autonomous driving to predict trajectories through learning from expert demonstrations, DNN-based methods suffer from data-scarcity, while IRL-based approaches often struggle with generalizability, making both hard... | [
"Reinforcement Learning",
"Trajectory Prediction",
"Autonomous Driving"
] | null | 217 | null | null |
LiwdXkMsDv | Uncertainty-Aware Decision Transformer for Stochastic Driving Environments | https://openreview.net/forum?id=LiwdXkMsDv | [
"Zenan Li",
"Fan Nie",
"Qiao Sun",
"Fang Da",
"Hang Zhao"
] | null | null | Offline Reinforcement Learning (RL) enables policy learning without active interactions, making it especially appealing for self-driving tasks. Recent successes of Transformers inspire casting offline RL as sequence modeling, which, however, fails in stochastic environments with incorrect assumptions that identical act... | [
"Self-Driving",
"Decision Transformer",
"Uncertainty-Aware Planning"
] | We present UNREST, an uncertainty-aware decision transformer to apply offline RL in stochastic driving environments. | 220 | 2309.16397 | title_snapshot |
3ZAgXBRvla | FlowBotHD: History-Aware Diffuser Handling Ambiguities in Articulated Objects Manipulation | https://openreview.net/forum?id=3ZAgXBRvla | [
"Yishu Li",
"Wen Hui Leng",
"Yiming Fang",
"Ben Eisner",
"David Held"
] | null | null | We introduce a novel approach to manipulate articulated objects with ambiguities, such as opening a door, in which multi-modality and occlusions create ambiguities about the opening side and direction. Multi-modality occurs when the method to open a fully closed door (push, pull, slide) is uncertain, or the side from w... | [
"Ambiguity",
"Multi-modality",
"Occlusion",
"Articulated Objects",
"Diffusion"
] | We present FlowBotHD, a history-aware diffuser that handles multi-modality and occlusions in articulated object manipulation | 222 | 2410.07078 | title_snapshot |
MwZJ96Okl3 | Modeling Drivers’ Situational Awareness from Eye Gaze for Driving Assistance | https://openreview.net/forum?id=MwZJ96Okl3 | [
"Abhijat Biswas",
"Pranay Gupta",
"Shreeya Khurana",
"David Held",
"Henny Admoni"
] | null | null | Intelligent driving assistance can alert drivers to objects in their environment; however, such systems require a model of drivers' situational awareness (SA) (what aspects of the scene they are already aware of) to avoid unnecessary alerts.
Moreover, collecting the data to train such an SA model is challenging:
bein... | [
"driver awareness",
"driving assistance",
"situational awareness"
] | We propose a new protocol to record drivers' situational awareness, use it to collect a dataset, and build a predictive SA model | 227 | null | null |
RMkdcKK7jq | SLR: Learning Quadruped Locomotion without Privileged Information | https://openreview.net/forum?id=RMkdcKK7jq | [
"Shiyi Chen",
"Zeyu Wan",
"Shiyang Yan",
"Chun Zhang",
"Weiyi Zhang",
"Qiang Li",
"Debing Zhang",
"Fasih Ud Din Farrukh"
] | null | null | Traditional reinforcement learning control for quadruped robots often relies on privileged information, demanding meticulous selection and precise estimation, thereby imposing constraints on the development process. This work proposes a Self-learning Latent Representation (SLR) method, which achieves high-performance c... | [
"Locomotion",
"Reinforcement Learning",
"Privileged Learning"
] | null | 228 | 2406.04835 | title_snapshot |
lyhS75loxe | A3VLM: Actionable Articulation-Aware Vision Language Model | https://openreview.net/forum?id=lyhS75loxe | [
"Siyuan Huang",
"Haonan Chang",
"Yuhan Liu",
"Yimeng Zhu",
"Hao Dong",
"Abdeslam Boularias",
"Peng Gao",
"Hongsheng Li"
] | null | null | Vision Language Models (VLMs) for robotics have received significant attention in recent years. As a VLM can understand robot observations and perform complex visual reasoning, it is regarded as a potential universal solution for general robotics challenges such as manipulation and navigation. However, previous robotic... | [
"LLM",
"VLM",
"Manipulation",
"Articulation"
] | We introudce the A3VLM, focusing on the articulation structure and action affordances of objects. | 229 | 2406.07549 | title_snapshot |
3i7j8ZPnbm | UMI-on-Legs: Making Manipulation Policies Mobile with Manipulation-Centric Whole-body Controllers | https://openreview.net/forum?id=3i7j8ZPnbm | [
"Huy Ha",
"Yihuai Gao",
"Zipeng Fu",
"Jie Tan",
"Shuran Song"
] | null | null | We introduce UMI-on-Legs, a new framework that combines real-world and simulation data for quadruped manipulation systems. We scale task-centric data collection in the real world using a handheld gripper (UMI), providing a cheap way to demonstrate task-relevant manipulation skills without a robot. Simultaneously, we s... | [
"Manipulation",
"Visuo-motor Policy",
"Whole-body Controller"
] | A framework for scalable quadruped manipulation, combining manipulation policies trained from real-world demonstrations with whole-body controllers trained from simulation. | 230 | 2407.10353 | title_snapshot |
wSWMsjuMTI | ManiWAV: Learning Robot Manipulation from In-the-Wild Audio-Visual Data | https://openreview.net/forum?id=wSWMsjuMTI | [
"Zeyi Liu",
"Cheng Chi",
"Eric Cousineau",
"Naveen Kuppuswamy",
"Benjamin Burchfiel",
"Shuran Song"
] | null | null | Audio signals provide rich information for the robot interaction and object properties through contact. These information can surprisingly ease the learning of contact-rich robot manipulation skills, especially when the visual information alone is ambiguous or incomplete. However, the usage of audio data in robot manip... | [
"Robot Manipulation",
"Imitation Learning",
"Audio"
] | We introduce ManiWAV: a data collection device to collect in-the-wild human demonstrations with synchronous audio and visual feedback, and a corresponding policy interface to learn robot manipulation policy directly from the demonstrations. | 231 | 2406.19464 | title_snapshot |
ySI0tBYxpz | Gaitor: Learning a Unified Representation Across Gaits for Real-World Quadruped Locomotion | https://openreview.net/forum?id=ySI0tBYxpz | [
"Alexander Luis Mitchell",
"Wolfgang Merkt",
"Aristotelis Papatheodorou",
"Ioannis Havoutis",
"Ingmar Posner"
] | null | null | The current state-of-the-art in quadruped locomotion is able to produce a variety of complex motions. These methods either rely on switching between a discrete set of skills or learn a distribution across gaits using complex black-box models. Alternatively, we present Gaitor, which learns a disentangled and 2D represen... | [
"Representation Learning",
"Learning for Control",
"Quadruped Control"
] | We create a structured latent representation across locomotion skills and show that continuous gait transitions emerge in the space automatically. | 233 | 2405.19452 | title_snapshot |
edP2dmingV | Large Scale Mapping of Indoor Magnetic Field by Local and Sparse Gaussian Processes | https://openreview.net/forum?id=edP2dmingV | [
"Iad ABDUL-RAOUF",
"Vincent Gay-Bellile",
"Cyril JOLY",
"Steve Bourgeois",
"Alexis Paljic"
] | null | null | Magnetometer-based indoor navigation uses variations in the magnetic field to determine the robot's location. For that, a magnetic map of the environment has to be built beforehand from a collection of localized magnetic measurements. Existing solutions built on sparse Gaussian Process (GP) regression do not scale well... | [
"Gaussian process regression",
"magnetic field maps",
"indoor localization"
] | We introduce a new large-scale magnetic field map model based on a combination of a sparse approximation of Gaussian process regression and a local expert aggregation method. | 236 | null | null |
r6ZhiVYriY | Trust the PRoC3S: Solving Long-Horizon Robotics Problems with LLMs and Constraint Satisfaction | https://openreview.net/forum?id=r6ZhiVYriY | [
"Aidan Curtis",
"Nishanth Kumar",
"Jing Cao",
"Tomás Lozano-Pérez",
"Leslie Pack Kaelbling"
] | null | null | Recent developments in pretrained large language models (LLMs) applied to robotics have demonstrated their capacity for sequencing a set of discrete skills to achieve open-ended goals in simple robotic tasks. In this paper, we examine the topic of LLM planning for a set of *continuously parameterized* skills whose exec... | [
"LLMs for planning",
"task and motion planning",
"constraint satisfaction"
] | having LLM's produce code that corresponds to a CSP, then solving that CSP separately, enables solving long-horizon robotic manipulation tasks with complex and realistic constratins | 237 | 2406.05572 | title_snapshot |
cNI0ZkK1yC | Flow as the Cross-domain Manipulation Interface | https://openreview.net/forum?id=cNI0ZkK1yC | [
"Mengda Xu",
"Zhenjia Xu",
"Yinghao Xu",
"Cheng Chi",
"Gordon Wetzstein",
"Manuela Veloso",
"Shuran Song"
] | null | null | We present Im2Flow2Act, a scalable learning framework that enables robots to acquire real-world manipulation skills without the need of real-world robot training data. The key idea behind Im2Flow2Act is to use object flow as the manipulation interface, bridging domain gaps between different embodiments (i.e., human and... | [
"Robots",
"Learning",
"cross-domain",
"cross-embodiment"
] | We present Im2Flow2Act, a scalable learning framework that enables robots to acquire manipulation skills from diverse data sources. | 238 | 2407.15208 | title_snapshot |
SW8ntpJl0E | JA-TN: Pick-and-Place Towel Shaping from Crumpled States based on TransporterNet with Joint-Probability Action Inference | https://openreview.net/forum?id=SW8ntpJl0E | [
"Halid Abdulrahim Kadi",
"Kasim Terzić"
] | null | null | Towel manipulation is a crucial step towards more general cloth manipulation. However, folding a towel from an arbitrarily crumpled state and recovering from a failed folding step remain critical challenges in robotics. We propose joint-probability action inference JA-TN, as a way to improve TransporterNet's operationa... | [
"Cloth Manipulation",
"Imitation Learning",
"Sim2Real Transfer"
] | Our JA-TN towel-shaping controller extends the TransporterNet architecture with our joint-probability action inference and achieves state-of-the-art performance for flattening, folding-from-flattened and folding-from-crumpled tasks in simulation. | 239 | null | null |
KdVLK0Wo5z | PoliFormer: Scaling On-Policy RL with Transformers Results in Masterful Navigators | https://openreview.net/forum?id=KdVLK0Wo5z | [
"Kuo-Hao Zeng",
"Zichen Zhang",
"Kiana Ehsani",
"Rose Hendrix",
"Jordi Salvador",
"Alvaro Herrasti",
"Ross Girshick",
"Aniruddha Kembhavi",
"Luca Weihs"
] | null | null | We present PoliFormer (Policy Transformer), an RGB-only indoor navigation agent trained end-to-end with reinforcement learning at scale that generalizes to the real-world without adaptation despite being trained purely in simulation. PoliFormer uses a foundational vision transformer encoder with a causal transformer de... | [
"Embodied Navigation",
"On-Policy RL",
"Transformer Policy"
] | null | 243 | 2406.20083 | title_snapshot |
jnubz7wB2w | Verification of Neural Control Barrier Functions with Symbolic Derivative Bounds Propagation | https://openreview.net/forum?id=jnubz7wB2w | [
"Hanjiang Hu",
"Yujie Yang",
"Tianhao Wei",
"Changliu Liu"
] | null | null | Control barrier functions (CBFs) are important in safety-critical systems and robot control applications. Neural networks have been used to parameterize and synthesize CBFs with bounded control input for complex systems. However, it is still challenging to verify pre-trained neural networks CBFs (neural CBFs) in an eff... | [
"Learning for control",
"control barrier function",
"formal verification"
] | We propose a new neural CBF verification framework using symbolic derivative bound propagation to give tight and efficient bounds for verifying forward invariance. | 253 | 2410.16281 | title_snapshot |
9jJP2J1oBP | Leveraging Mutual Information for Asymmetric Learning under Partial Observability | https://openreview.net/forum?id=9jJP2J1oBP | [
"Hai Huu Nguyen",
"Long Dinh Van The",
"Christopher Amato",
"Robert Platt"
] | null | null | Even though partial observability is prevalent in robotics, most reinforcement learning studies avoid it due to the difficulty of learning a policy that can efficiently memorize past events and seek information. Fortunately, in many cases, learning can be done in an asymmetric setting where states are available during ... | [
"Partial Observability",
"Mutual Information",
"Reinforcement Learning"
] | This paper proposes to use the mutual information between the history and the state to improve learning under partial observability, assuming state availability during training. | 256 | null | null |
s0vHSq5QEv | Generalizing End-To-End Autonomous Driving In Real-World Environments Using Zero-Shot LLMs | https://openreview.net/forum?id=s0vHSq5QEv | [
"Zeyu Dong",
"Yimin Zhu",
"Yansong Li",
"Kevin Mahon",
"Yu Sun"
] | null | null | Traditional autonomous driving methods adopt modular design, decomposing tasks into sub-tasks, including perception, prediction, planning, and control. In contrast, end-to-end autonomous driving directly outputs actions from raw sensor data, avoiding error accumulation. However, training an end-to-end model requires a ... | [
"End-to-end Autonomous Driving",
"Large Vision-Language Model",
"Generalization"
] | We proposed an architecture that combines LLMs with end-to-end driving models, characterized by low data requirements, strong generalization ability, and suitability for real-world closed-loop inference. | 257 | 2411.14256 | title_snapshot |
Isp19rFFV4 | Multi-Strategy Deployment-Time Learning and Adaptation for Navigation under Uncertainty | https://openreview.net/forum?id=Isp19rFFV4 | [
"Abhishek Paudel",
"Xuesu Xiao",
"Gregory J. Stein"
] | null | null | We present an approach for performant point-goal navigation in unfamiliar partially-mapped environments. When deployed, our robot runs multiple strategies for deployment-time learning and visual domain adaptation in parallel and quickly selects the best-performing among them. Choosing between policies as they are learn... | [
"policy selection",
"domain adaptation",
"navigation under uncertainty"
] | We present an approach that enables a robot to deploy multiple learning/adaptation strategies during deployment and pick the best one. | 258 | null | null |
xeFKtSXPMd | OCCAM: Online Continuous Controller Adaptation with Meta-Learned Models | https://openreview.net/forum?id=xeFKtSXPMd | [
"Hersh Sanghvi",
"Spencer Folk",
"Camillo Jose Taylor"
] | null | null | Control tuning and adaptation present a significant challenge to the usage of robots in diverse environments. It is often nontrivial to find a single set of control parameters by hand that work well across the broad array of environments and conditions that a robot might encounter. Automated adaptation approaches must ... | [
"Controller Adaptation",
"Robot Model Learning",
"Meta-Learning"
] | We present a framework for online controller adaptation that learns prior predictive models of system performance that quickly adapt to online data and can be used to to optimize control parameters for diverse physical and simulated robots. | 261 | 2406.17620 | title_snapshot |
TzqKmIhcwq | Structured Bayesian Meta-Learning for Data-Efficient Visual-Tactile Model Estimation | https://openreview.net/forum?id=TzqKmIhcwq | [
"Shaoxiong Yao",
"Yifan Zhu",
"Kris Hauser"
] | null | null | Estimating visual-tactile models of deformable objects is challenging because vision suffers from occlusion, while touch data is sparse and noisy. We propose a novel data-efficient method for dense heterogeneous model estimation by leveraging experience from diverse training objects. The method is based on Bayesian M... | [
"Multimodal perception",
"tactile sensing",
"few-shot learning"
] | Our SBML framework enables data-efficient visual-tactile model estimation by learning a prior of visual-tactile model from diverse real world objects. | 262 | null | null |
CPQW5kc0pe | VoxAct-B: Voxel-Based Acting and Stabilizing Policy for Bimanual Manipulation | https://openreview.net/forum?id=CPQW5kc0pe | [
"I-Chun Arthur Liu",
"Sicheng He",
"Daniel Seita",
"Gaurav S. Sukhatme"
] | null | null | Bimanual manipulation is critical to many robotics applications. In contrast to single-arm manipulation, bimanual manipulation tasks are challenging due to higher-dimensional action spaces. Prior works leverage large amounts of data and primitive actions to address this problem, but may suffer from sample inefficiency ... | [
"bimanual manipulation",
"voxel representation",
"vision language models"
] | We propose VoxAct-B for bimanual manipulation, a language-conditioned, voxel-based method that leverages Vision Language Models to prioritize key regions within the scene and subsequently reconstruct a voxel grid with increased voxel resolution. | 264 | 2407.04152 | title_snapshot |
9HkElMlPbU | Contrastive Imitation Learning for Language-guided Multi-Task Robotic Manipulation | https://openreview.net/forum?id=9HkElMlPbU | [
"Teli Ma",
"Jiaming Zhou",
"Zifan Wang",
"Ronghe Qiu",
"Junwei Liang"
] | null | null | Developing robots capable of executing various manipulation tasks, guided by natural language instructions and visual observations of intricate real-world environments, remains a significant challenge in robotics. Such robot agents need to understand linguistic commands and distinguish between the requirements of diff... | [
"Contrastive imitation learning",
"Multi-task learning",
"Robotic manipulation"
] | null | 269 | 2406.09738 | title_snapshot |
End of preview. Expand in Data Studio
README.md exists but content is empty.
- Downloads last month
- 55