Data-Bootstrapped, Physics-Informed Framework for Object Rearrangement

Object rearrangement, which involves arranging objects step-by-step to achieve tidy states, is critical in robotic applications. Progress in this area is often constrained by issues such as high-cost data collection and physically infeasible trajectory prediction. To address these challenges, we propose the Data-Bootstrapped, Physics-Informed Rearrangement (DPR) framework, which leverages a transformer for sequential decision making. Specifically, DPR integrates Enhanced Data Generation with a Physics Reward Feedback Transformer. Enhanced Data Generation consists of Random Trajectory Reverse for producing high-quality training data and Bootstrapped Trajectory Synthesis, which leverages the transformer’s sequence modeling to diversify training trajectories. To ensure the feasibility of the generated trajectories and to improve the transformer’s performance, we incorporate a Physical Reward Feedback mechanism into the transformer. Experiments on ball and room rearrangement tasks show that DPR significantly outperforms existing methods in terms of both efficiency and effectiveness. Code will be released soon.

Authors

• Alex Wong
• Zhiwei Dong

Keywords

• Training
• Decision making
• Training data
• Data collection
• Transformers
• Robustness
• Data models
• Trajectory
• Physics
• Intelligent robots
• Object Rearrangement
• High-quality Training
• Training Trajectories
• Extensive Data
• Multiple Objects
• Physical Constraints
• Transformer Model
• Training Examples
• Reward Function
• Past Conditions
• Markov Decision Process
• Collision Frequency
• Real Scenes
• Entire Trajectory
• Limited Training Data
• Trajectory Generation
• Coverage Score
• Transformer-based Methods
• Temporal Difference Learning
• Trajectory Dataset
• Offline Learning
• Discrete State Space