Experimental investigation of pose informed reinforcement learning for skid-steered visual navigation
This addresses the problem of deploying autonomous navigation for skid-steered vehicles in off-road settings, where modeling complexities hinder automation, representing an incremental advance in this domain.
The paper tackles the challenge of automating skid-steered vehicles for visual navigation by proposing a structured learning approach, which shows significantly improved performance in simulations and hardware evaluations compared to existing methods.
Vision-based lane keeping is a topic of significant interest in the robotics and autonomous ground vehicles communities in various on-road and off-road applications. The skid-steered vehicle architecture has served as a useful vehicle platform for human controlled operations. However, systematic modeling, especially of the skid-slip wheel terrain interactions (primarily in off-road settings) has created bottlenecks for automation deployment. End-to-end learning based methods such as imitation learning and deep reinforcement learning, have gained prominence as a viable deployment option to counter the lack of accurate analytical models. However, the systematic formulation and subsequent verification/validation in dynamic operation regimes (particularly for skid-steered vehicles) remains a work in progress. To this end, a novel approach for structured formulation for learning visual navigation is proposed and investigated in this work. Extensive software simulations, hardware evaluations and ablation studies now highlight the significantly improved performance of the proposed approach against contemporary literature.