ashiqali98 commited on
Commit
5b8bf46
·
verified ·
1 Parent(s): 14fd033

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -27
README.md CHANGED
@@ -10,21 +10,19 @@ tags:
10
  library_name: pytorch
11
  ---
12
 
13
- # SDPC Diffusion Policy — Quadrotor Corridor Navigation
14
 
15
  Trained checkpoint for **SDPC (Safe Diffusion Policy with Constraint)**, an image-(and goal-)conditioned
16
  trajectory diffusion policy for quadrotor obstacle avoidance, with hard constraints enforced at
17
  inference time via an SLSQP projection step. Developed as part of a master's thesis at Paderborn
18
  University.
19
 
20
- Code: [ashiqlathief/dpcc-thesis1](https://github.com/ashiqlathief/dpcc-thesis1)
21
 
22
  ## What this checkpoint is
23
 
24
- A single trained run — `H8_K20_Dmodels.ImagePoseCondUNet1DTemporalCondModel_Evitp_L384`, seed `7` —
25
- i.e. the `ImagePoseCondUNet1DTemporalCondModel` denoiser (UNet + ViT-P image encoder, goal/pose
26
- conditioned). It predicts an 8-step action-chunk trajectory (`horizon=8`) via a 20-step DDPM
27
- denoising process (`n_diffusion_steps=20`), conditioned on 1 past FPV observation (`n_obs_steps=1`)
28
  and the relative goal vector.
29
 
30
  The policy was trained on FPV-camera demonstrations of a quadrotor navigating a corridor around
@@ -83,10 +81,8 @@ works") for that mechanism.
83
 
84
  ## How to use
85
 
86
- This checkpoint is meant to be loaded with the training/eval code in
87
- [ashiqlathief/dpcc-thesis1](https://github.com/ashiqlathief/dpcc-thesis1), not standalone —
88
- `model_config.pkl`/`diffusion_config.pkl` reference model classes defined in that repo's
89
- `diffuser/models/` package.
90
 
91
  ```bash
92
  # from the dpcc-thesis1 repo root, with env_isaaclab active
@@ -102,20 +98,4 @@ import diffuser.utils as utils
102
  RUN_DIR = "isaac/logs/avoiding-crazyflie/diffusion/H8_K20_Dmodels.ImagePoseCondUNet1DTemporalCondModel_Evitp_L384/7"
103
  diff_exp = utils.load_diffusion(RUN_DIR, epoch="best", device="cuda:0")
104
  diffusion = diff_exp.diffusion.eval()
105
- ```
106
-
107
- ## Limitations
108
-
109
- - Trained and evaluated entirely in simulation on a fixed, small set of static obstacle layouts —
110
- not validated for arbitrary real-world scenes.
111
- - No dynamic-obstacle-avoidance guarantee: obstacle positions used at inference are either
112
- ground-truth (simulation) or perceived via a depth-camera pipeline with known accuracy limits
113
- (see the repo's `depth_obstacle_estimator.py` and its documented bug history).
114
- - Quantitative success-rate / constraint-violation results across projection variants are reported
115
- in the accompanying thesis, not restated here to avoid duplicating numbers that may be revised
116
- before submission.
117
-
118
- ## Citation
119
-
120
- If you use this checkpoint, please cite the accompanying master's thesis (Paderborn University,
121
- 2026) — citation details to be added on publication.
 
10
  library_name: pytorch
11
  ---
12
 
13
+ # SDPC Diffusion Policy for Quadrotor Corridor Navigation
14
 
15
  Trained checkpoint for **SDPC (Safe Diffusion Policy with Constraint)**, an image-(and goal-)conditioned
16
  trajectory diffusion policy for quadrotor obstacle avoidance, with hard constraints enforced at
17
  inference time via an SLSQP projection step. Developed as part of a master's thesis at Paderborn
18
  University.
19
 
20
+ Code: [ashiqlathief/SDPC-imagepolicy](https://github.com/ashiqlathief/SDPC-imagepolicy)
21
 
22
  ## What this checkpoint is
23
 
24
+ A single trained run `H8_K20_Dmodels.ImagePoseCondUNet1DTemporalCondModel_Evitp_L384`, seed `7` i.e. the `ImagePoseCondUNet1DTemporalCondModel` denoiser (UNet + ViT-P image encoder, goal/pose
25
+ conditioned). It predicts an 8-step action-chunk trajectory (`horizon=8`) via a 20-step DDPM denoising process (`n_diffusion_steps=20`), conditioned on 1 past FPV observation (`n_obs_steps=1`)
 
 
26
  and the relative goal vector.
27
 
28
  The policy was trained on FPV-camera demonstrations of a quadrotor navigating a corridor around
 
81
 
82
  ## How to use
83
 
84
+ This checkpoint is meant to be loaded with the training/eval code in [ashiqlathief/SDPC-imagepolicy](https://github.com/ashiqlathief/SDPC-imagepolicy), not standalone and
85
+ `model_config.pkl`/`diffusion_config.pkl` reference model classes defined in that repo's `diffuser/models/` package.
 
 
86
 
87
  ```bash
88
  # from the dpcc-thesis1 repo root, with env_isaaclab active
 
98
  RUN_DIR = "isaac/logs/avoiding-crazyflie/diffusion/H8_K20_Dmodels.ImagePoseCondUNet1DTemporalCondModel_Evitp_L384/7"
99
  diff_exp = utils.load_diffusion(RUN_DIR, epoch="best", device="cuda:0")
100
  diffusion = diff_exp.diffusion.eval()
101
+ ```