# ============================================================ # LumiAI - LIVE AI STREET LIGHTING DASHBOARD # ============================================================ import streamlit as st import pandas as pd import numpy as np import plotly.graph_objects as go from datetime import datetime import time # ============================================================ # PAGE CONFIGURATION # ============================================================ st.set_page_config( page_title="LumiAI Smart Street Lighting", page_icon="πŸ’‘", layout="wide" ) # ============================================================ # TITLE # ============================================================ st.title("πŸ’‘ LumiAI") st.subheader( "AI-Based Adaptive Smart Street Lighting " "for Energy Optimization" ) st.markdown( "### YOLO + LSTM + Q-Learning" ) # ============================================================ # SIDEBAR # ============================================================ st.sidebar.header("System Control") road = st.sidebar.selectbox( "Select Road", [ "Road A", "Road B", "Road C", "Road D" ] ) auto_refresh = st.sidebar.checkbox( "Live Refresh", value=True ) # ============================================================ # SIMULATED LIVE YOLO DATA # ============================================================ # In final system these values will come from YOLO vehicle_count = np.random.randint( 5, 100 ) pedestrian_count = np.random.randint( 0, 20 ) # ============================================================ # TRAFFIC CLASSIFICATION # ============================================================ if vehicle_count < 20: traffic_level = "LOW" elif vehicle_count < 60: traffic_level = "MEDIUM" else: traffic_level = "HIGH" # ============================================================ # LSTM TRAFFIC PREDICTION # ============================================================ # Demonstration prediction # Replace this with actual LSTM prediction predicted_traffic = int( vehicle_count + np.random.randint(-5, 15) ) predicted_traffic = max( 0, predicted_traffic ) # ============================================================ # Q-LEARNING OPTIMAL BRIGHTNESS # ============================================================ if predicted_traffic >= 60: brightness = 100 elif predicted_traffic >= 20: brightness = 70 elif predicted_traffic > 0: brightness = 50 else: brightness = 30 # ============================================================ # ENERGY MODEL # ============================================================ MAX_POWER = 100 # Watts power = ( MAX_POWER * brightness / 100 ) # 15-minute energy consumption energy = ( power * 0.25 / 1000 ) # Conventional energy conventional_energy = ( MAX_POWER * 0.25 / 1000 ) energy_saved = ( conventional_energy - energy ) saving_percentage = ( energy_saved / conventional_energy ) * 100 # ============================================================ # CURRENT TIME # ============================================================ current_time = datetime.now().strftime( "%H:%M:%S" ) # ============================================================ # TOP METRICS # ============================================================ col1, col2, col3, col4, col5 = st.columns(5) with col1: st.metric( "πŸš— Vehicles", vehicle_count ) with col2: st.metric( "🚢 Pedestrians", pedestrian_count ) with col3: st.metric( "πŸ“Š Traffic", traffic_level ) with col4: st.metric( "πŸ’‘ Brightness", f"{brightness}%" ) with col5: st.metric( "⚑ Energy Saving", f"{saving_percentage:.1f}%" ) st.divider() # ============================================================ # SYSTEM STATUS # ============================================================ col1, col2, col3 = st.columns(3) with col1: st.success( "🟒 AI SYSTEM ACTIVE" ) with col2: st.info( f"πŸ•’ Last Update: {current_time}" ) with col3: st.success( f"πŸ“ {road}" ) # ============================================================ # TRAFFIC + PREDICTION # ============================================================ st.header("🚦 Traffic Intelligence") col1, col2 = st.columns(2) with col1: st.subheader( "Current vs Predicted Traffic" ) traffic_data = pd.DataFrame({ "Time": [ "Now", "+15 min" ], "Vehicles": [ vehicle_count, predicted_traffic ] }) fig = go.Figure() fig.add_trace( go.Bar( x=traffic_data["Time"], y=traffic_data["Vehicles"], name="Vehicles" ) ) fig.update_layout( yaxis_title="Vehicle Count", xaxis_title="Time", height=350 ) st.plotly_chart( fig, use_container_width=True ) with col2: st.subheader( "AI Lighting Decision" ) lighting_data = pd.DataFrame({ "Parameter": [ "Traffic", "Predicted Traffic", "Brightness", "Power" ], "Value": [ vehicle_count, predicted_traffic, f"{brightness}%", f"{power:.1f} W" ] }) st.table( lighting_data ) # ============================================================ # BRIGHTNESS GAUGE # ============================================================ st.header("πŸ’‘ Adaptive Lighting") fig = go.Figure( go.Indicator( mode="gauge+number", value=brightness, title={ "text": "Optimal Brightness (%)" }, gauge={ "axis": { "range": [0, 100] } } ) ) fig.update_layout( height=350 ) st.plotly_chart( fig, use_container_width=True ) # ============================================================ # ENERGY ANALYTICS # ============================================================ st.header("⚑ Energy Optimization") col1, col2, col3 = st.columns(3) with col1: st.metric( "Conventional", f"{conventional_energy:.4f} kWh" ) with col2: st.metric( "LumiAI", f"{energy:.4f} kWh" ) with col3: st.metric( "Saved", f"{energy_saved:.4f} kWh" ) # ============================================================ # ENERGY COMPARISON GRAPH # ============================================================ energy_df = pd.DataFrame({ "System": [ "Conventional", "LumiAI" ], "Energy": [ conventional_energy, energy ] }) fig = go.Figure() fig.add_trace( go.Bar( x=energy_df["System"], y=energy_df["Energy"], name="Energy" ) ) fig.update_layout( title="Energy Consumption Comparison", yaxis_title="Energy (kWh)", height=400 ) st.plotly_chart( fig, use_container_width=True ) # ============================================================ # AI DECISION EXPLANATION # ============================================================ st.header("🧠 AI Decision") if brightness == 100: st.error( "HIGH TRAFFIC β†’ Maximum brightness selected " "for safety." ) elif brightness == 70: st.warning( "MEDIUM TRAFFIC β†’ 70% brightness selected." ) elif brightness == 50: st.info( "LOW TRAFFIC β†’ 50% brightness selected." ) else: st.success( "VERY LOW TRAFFIC β†’ Minimum safe brightness." ) # ============================================================ # ROAD STATUS # ============================================================ st.header("πŸ—ΊοΈ Road Network Status") road_data = pd.DataFrame({ "Road": [ "Road A", "Road B", "Road C", "Road D" ], "Traffic": [ np.random.randint(10, 100), np.random.randint(10, 100), np.random.randint(10, 100), np.random.randint(10, 100) ] }) road_data["Brightness"] = road_data[ "Traffic" ].apply( lambda x: 100 if x >= 60 else 70 if x >= 20 else 50 ) st.dataframe( road_data, use_container_width=True ) # ============================================================ # FOOTER # ============================================================ st.divider() st.caption( "LumiAI | YOLO + LSTM + Q-Learning | " "AI-Based Energy Optimization" ) # ============================================================ # AUTO REFRESH # ============================================================ if auto_refresh: time.sleep(3) st.rerun()