Risk-Aware Obstacle Avoidance Algorithm for Real-Time Applications

TL;DR

Robust autonomous navigation for autonomous surface vessels under uncertainty is addressed by a hybrid risk-aware framework that fuses probabilistic risk maps, built from a Bayesian Belief Network (BBN), with smooth trajectory optimization. The RA-RRT* planner uses the BBN-derived hazard probabilities to bias planning decisions and avoid high-risk regions, with trajectories refined by B-spline smoothing. Three planning modes are explored: shortest-path, risk-minimizing, and an α-weighted compromise that balances path length and safety via the cost $J(p)=\alpha D(p)+(1-\alpha)\beta B(p)$. Results show reduced cumulative hazardous-event probability and smoother trajectories in dynamic environments compared with geometry-only methods, demonstrating improved safety and autonomy in uncertain marine settings. The work lays a foundation for real-time, risk-aware autonomous navigation and highlights future avenues such as more risk-influencing factors, platform validation, and multi-sensor integration.

Abstract

Robust navigation in changing marine environments requires autonomous systems capable of perceiving, reasoning, and acting under uncertainty. This study introduces a hybrid risk-aware navigation architecture that integrates probabilistic modeling of obstacles along the vehicle path with smooth trajectory optimization for autonomous surface vessels. The system constructs probabilistic risk maps that capture both obstacle proximity and the behavior of dynamic objects. A risk-biased Rapidly Exploring Random Tree (RRT) planner leverages these maps to generate collision-free paths, which are subsequently refined using B-spline algorithms to ensure trajectory continuity. Three distinct RRT* rewiring modes are implemented based on the cost function: minimizing the path length, minimizing risk, and optimizing a combination of the path length and total risk. The framework is evaluated in experimental scenarios containing both static and dynamic obstacles. The results demonstrate the system's ability to navigate safely, maintain smooth trajectories, and dynamically adapt to changing environmental risks. Compared with conventional LIDAR or vision-only navigation approaches, the proposed method shows improvements in operational safety and autonomy, establishing it as a promising solution for risk-aware autonomous vehicle missions in uncertain and dynamic environments.

Figures (5)

• Figure 1: System architecture.
• Figure 2: Path planning result as the dynamic obstacle moves towards the center of the search space. The path is computed with $\alpha=0.2$. Path is replanned every $10$ seconds.
• Figure 3: The rewiring path results in different value of weighting-coefficient ($\alpha$).
• Figure 4: Box plots of path length $D(p^*)$ and hazardous event probability $\frac{1}{n} \sum \Pr\{E_i\}$ for different values of $\alpha$.
• Figure :