Learning When to Jump for Off-road Navigation
🤖 AI Summary
This work addresses the limitations of traditional off-road navigation methods, which often neglect vehicle dynamics and struggle with complex terrains requiring dynamic maneuvers such as jumping. To overcome this, the authors propose Motion-Aware Traversability (MAT), a novel representation that models terrain traversability as a velocity-dependent Gaussian function. By integrating a deep learning-based perception model with a two-stage online planning architecture, the system predicts Gaussian parameters in real time and dynamically updates terrain costs according to current motion constraints. This approach transcends the conventional scalar traversability scores by explicitly enabling dynamic behaviors like jumping. Experimental results demonstrate that the system operates in real time in both simulation and real-world environments, reducing detour paths by 75% while ensuring safe and efficient navigation across challenging terrains.
Application Category
📝 Abstract
Low speed does not always guarantee safety in off-road driving. For instance, crossing a ditch may be risky at a low speed due to the risk of getting stuck, yet safe at a higher speed with a controlled, accelerated jump. Achieving such behavior requires path planning that explicitly models complex motion dynamics, whereas existing methods often neglect this aspect and plan solely based on positions or a fixed velocity. To address this gap, we introduce Motion-aware Traversability (MAT) representation to explicitly model terrain cost conditioned on actual robot motion. Instead of assigning a single scalar score for traversability, MAT models each terrain region as a Gaussian function of velocity. During online planning, we decompose the terrain cost computation into two stages: (1) predict terrain-dependent Gaussian parameters from perception in a single forward pass, (2) efficiently update terrain costs for new velocities inferred from current dynamics by evaluating these functions without repeated inference. We develop a system that integrates MAT to enable agile off-road navigation and evaluate it in both simulated and real-world environments with various obstacles. Results show that MAT achieves real-time efficiency and enhances the performance of off-road navigation, reducing path detours by 75% while maintaining safety across challenging terrains.
Problem
Research questions and friction points this paper is trying to address.
off-road navigation
motion dynamics
traversability
path planning
terrain cost
Innovation
Methods, ideas, or system contributions that make the work stand out.
Motion-aware Traversability
off-road navigation
velocity-dependent traversability
Gaussian terrain modeling
real-time path planning
🔎 Similar Papers
2024-09-28arXiv.orgCitations: 0
