cleaning
AI-Integrated Autonomous Robotics for Solar Panel Cleaning and Predictive Maintenance
Built by Not specified
Overview
This study proposes an AI-integrated autonomous robotic system combining real-time monitoring, predictive analytics, and intelligent cleaning for enhanced solar panel performance. The system integrates CNN-LSTM-based fault detection, Reinforcement Learning (DQN)-driven robotic cleaning, and Edge AI analytics for low-latency decision-making. Thermal and LiDAR-equipped drones detect panel faults, while ground robots clean panel surfaces based on real-time dust and temperature data. The system achieved an average cleaning efficiency of 91.3%, reducing dust density from 3.9 to 0.28 mg/m³, and restoring up to 31.2% energy output on heavily soiled panels.
Key facts
- Method
- scrubbing
- Navigation
- lidar_slam, vision, hybrid
Detailed specifications
Motion & kinematics1
- Fault Detection Accuracy
- 92.3%
Power & battery1
- Energy Restoration
- up to 31.2%
Other12
- Applications
- industrial_cleaning,disinfection
- Sub Category
- solar_panel_cleaning
- Dust Reduction
- 3.9 to 0.28 mg/m³
- Cleaning Method
- scrubbing
- Edge Ai Latency
- 47.2 ms
- Navigation Type
- lidar_slam,vision,hybrid
- Deployment Notes
- Deployed at Sitapura, Jaipur, India for a 72-hour field test.
- Industries Served
- solar_farms
- Obstacle Avoidance
- true
- Availability Status
- research-only
- Cleaning Efficiency
- 91.3%
- Additional Information
- - Achieves 91.3% cleaning efficiency with RL-based optimization. - Reduces dust density from 3.9 to 0.28 mg/m³. - Restores up to 31.2% energy output on heavily soiled panels. - Uses CNN-LSTM for fault detection with 92.3% accuracy. - Edge AI reduces latency by 63% compared to cloud processing. - Deployed in Sitapura, Jaipur for field testing.
Reviews for AI-Integrated Autonomous Robotics for Solar Panel Cleaning and Predictive Maintenance
Loading reviews…