Shadow Program Inversion with Differentiable Planning: A Framework for Unified Robot Program Parameter and Trajectory Optimization

2024-09-13

arXiv.org

AI Summary

This work addresses the challenge of jointly optimizing high-level task objectives—such as cycle time and trajectory smoothness—with low-level motion constraints—including collision avoidance and joint feasibility—in robotic program optimization. We propose an end-to-end differentiable program optimization framework. Its core innovation is the first integration of a differentiable collision-free motion planner (DGPMP2-ND) into the gradient-based optimization of parameterized robotic programs, enabling exact gradient propagation through motion constraints via implicit differentiation. Crucially, the framework preserves program semantics, ensuring human interpretability, editability, and formal verifiability. Experiments in household service and industrial assembly scenarios demonstrate significant reductions in task cycle time, zero collision rate, and full adherence to kinematic and geometric constraints—while maintaining program transparency and verification capability.

Technology Category

• Intelligent Robots: Learning & Optimization for ROB
• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Constraint Optimization

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

This paper presents SPI-DP, a novel first-order optimizer capable of optimizing robot programs with respect to both high-level task objectives and motion-level constraints. To that end, we introduce DGPMP2-ND, a differentiable collision-free motion planner for serial N-DoF kinematics, and integrate it into an iterative, gradient-based optimization approach for generic, parameterized robot program representations. SPI-DP allows first-order optimization of planned trajectories and program parameters with respect to objectives such as cycle time or smoothness subject to e.g. collision constraints, while enabling humans to understand, modify or even certify the optimized programs. We provide a comprehensive evaluation on two practical household and industrial applications.

Problem

• Optimizes robot program parameters
• Integrates collision-free motion planning
• Enables human-understandable program modifications

Innovation

• Differentiable Planning Framework
• First-order Optimizer SPI-DP
• Differentiable Collision-free Motion Planner