From Kinematics to Dynamics: Learning to Refine Hybrid Plans for Physically Feasible Execution

TL;DR

This paper presents a reinforcement learning approach that refines hybrid plans in robotics to ensure physically feasible trajectories respecting true dynamic constraints.

Contribution

It introduces a method that explicitly incorporates second-order dynamics constraints into reinforcement learning to improve plan feasibility in robotic tasks.

Findings

The approach reliably recovers physical feasibility of plans.

It effectively bridges the gap between initial plans and real dynamic execution.

The method outperforms traditional linear dynamics models in plan refinement.

Abstract

In many robotic tasks, agents must traverse a sequence of spatial regions to complete a mission. Such problems are inherently mixed discrete-continuous: a high-level action sequence and a physically feasible continuous trajectory. The resulting trajectory and action sequence must also satisfy problem constraints such as deadlines, time windows, and velocity or acceleration limits. While hybrid temporal planners attempt to address this challenge, they typically model motion using linear (first-order) dynamics, which cannot guarantee that the resulting plan respects the robot's true physical constraints. Consequently, even when the high-level action sequence is fixed, producing a dynamically feasible trajectory becomes a bi-level optimization problem. We address this problem via reinforcement learning in continuous space. We define a Markov Decision Process that explicitly incorporates analytical second-order constraints and use it to refine first-order plans generated by a hybrid planner. Our results show that this approach can reliably recover physical feasibility and effectively bridge the gap between a planner's initial first-order trajectory and the dynamics required for real execution.