A Risk-aware Planning Framework of UGVs in Off-Road Environment

TL;DR

This work tackles risk-aware motion planning for UGVs in off-road environments by integrating a risk assessment module based on artificial potential fields with a global–local planning framework. The global planner uses a novel Coarse2fine A* algorithm that operates on uncertainty maps derived from APF and is complemented by a quadratic programming-based path smoothing, while the local planner employs deterministic sampling in the Frenet frame to generate safe, feasible trajectories. The main contributions are the unified APF risk model for static and dynamic sources, the Coarse2fine A* approach with hierarchical maps and parallel search, and a rolling, constraint-aware local planning pipeline that achieves substantial safety improvements (over 30% reduction in path uncertainty) and real-time performance on hardware. The framework is validated through simulations and real-world UGV experiments, demonstrating practical feasibility for safe autonomous off-road operation and providing a foundation for further enhancements in uncertainty handling and emergency fallback strategies.

Abstract

Planning module is an essential component of intelligent vehicle study. In this paper, we address the risk-aware planning problem of UGVs through a global-local planning framework which seamlessly integrates risk assessment methods. In particular, a global planning algorithm named Coarse2fine A* is proposed, which incorporates a potential field approach to enhance the safety of the planning results while ensuring the efficiency of the algorithm. A deterministic sampling method for local planning is leveraged and modified to suit off-road environment. It also integrates a risk assessment model to emphasize the avoidance of local risks. The performance of the algorithm is demonstrated through simulation experiments by comparing it with baseline algorithms, where the results of Coarse2fine A* are shown to be approximately 30% safer than those of the baseline algorithms. The practicality and effectiveness of the proposed planning framework are validated by deploying it on a real-world system consisting of a control center and a practical UGV platform.

Figures (18)

• Figure 1: The framework of proposed system.
• Figure 2: The potential field map for static risk sources.
• Figure 3: The static field, dynamic field, and overall potential field of moving objects within a local area.
• Figure 4: The uncertainty maps: (a) and (c) are fine map and coarse map of the designed scenario; (b) and (d) are fine map and coarse map with randomly generated uncertainty.
• Figure 5: The visualization of the Coarse2fine A* algorithm.