Path Optimization for Ground Vehicles in Off-Road Terrain

TL;DR

This paper introduces an actuator space-based path optimization method for off-road ground vehicles, enabling high-speed, kinematically feasible path planning through gradient descent, validated in simulation and real-world tests.

Contribution

The paper proposes a novel actuator space approach that simplifies kinematic constraints into steering angle limits for efficient path optimization.

Findings

Paths are kinematically feasible and optimized for high-speed off-road travel.

Method successfully tested in simulation and on real vehicle at 5 m/s.

Gradient descent effectively finds optimal steering angle sequences.

Abstract

We present a method for path optimization for ground vehicles in off-road environments at high speeds. This path optimization considers the kinematic constraints of the vehicle. By thinking in the actuator space we can represent these constraints as limits in the space rather than derived properties of the path. In this paper we present a actuator space approach to path optimization for off-road ground vehicles. This is done by representing and operation on the path as a list of steering angles over the path length. This transforms the set of kinematic constraints into constraints on the steering angle. We then put this path into a gradient descent solver. This produced paths that are kinematically feasible and optimized in accordance with our cost function. Finally, we tested the system both in simulation and on an off-road vehicle at speeds of 5 m/s.