Data-Efficient Learning of High-Quality Controls for Kinodynamic Planning used in Vehicular Navigation

TL;DR

This paper introduces a learning framework that enhances kinodynamic motion planning for vehicles by efficiently identifying high-quality controls, leading to better paths with less computation across different environments.

Contribution

It presents a novel, data-efficient learning approach for control selection in kinodynamic planning, adaptable to various environments and integrated with biased and medial axis-based expansion strategies.

Findings

Improved path quality over random control methods.

Reduced computation time and fewer iterations needed.

Effective across different vehicle dynamics models.

Abstract

This paper aims to improve the path quality and computational efficiency of kinodynamic planners used for vehicular systems. It proposes a learning framework for identifying promising controls during the expansion process of sampling-based motion planners for systems with dynamics. Offline, the learning process is trained to return the highest-quality control that reaches a local goal state (i.e., a waypoint) in the absence of obstacles from an input difference vector between its current state and a local goal state. The data generation scheme provides bounds on the target dispersion and uses state space pruning to ensure high-quality controls. By focusing on the system's dynamics, this process is data efficient and takes place once for a dynamical system, so that it can be used for different environments with modular expansion functions. This work integrates the proposed learning process with a) an exploratory expansion function that generates waypoints with biased coverage over the reachable space, and b) proposes an exploitative expansion function for mobile robots, which generates waypoints using medial axis information. This paper evaluates the learning process and the corresponding planners for a first and second-order differential drive systems. The results show that the proposed integration of learning and planning can produce better quality paths than kinodynamic planning with random controls in fewer iterations and computation time.