Radio-Coverage-Aware Path Planning for Cooperative Autonomous Vehicles

TL;DR

This work addresses radio-coverage-aware cooperative mapping and path planning for fleets of autonomous vehicles that offload heavy computer vision tasks to edge or cloud servers. It introduces radio-weighted cost functions integrated into Dijkstra and A* search, framed as a Pareto-optimal multi-objective problem, and develops two scalable solvers: Weighted Dijkstra and Weighted A*. The approach leverages obstacle maps built via cooperative mapping and radio-weight maps derived from access points to maximize connectivity with minimal path-length increase, achieving obstacle-map accuracy below $2\%$ in simulations and notable improvements in radio coverage. The results demonstrate the practicality of integrating wireless constraints into planning for fog-enabled autonomous fleets in smart-city scenarios, enabling reliable offloading and safer navigation.

Abstract

Fleets of autonomous vehicles (AV) often are at the core of intelligent transportation scenarios for smart cities, and may require a wireless Internet connection to offload computer vision tasks to data centers located either in the edge or the cloud section of the network. Cooperation among AVs is successful when the environment is unknown, or changes dynamically, so as to improve coverage and trip time, and minimize the traveled distance. The AVs, while mapping the environment with range-based sensors, move across the wireless coverage areas, with consequences on the achieved access bit rate, latency, and handover rate. In this paper, we propose to modify the cost of path planning algorithms such as Dijkstra and A*, so that not only the traveled distance is considered in the best path solution, but also the radio coverage experience. To this aim, several radio-related cost-weighting functions are introduced and tested, to assess the performance of the proposed techniques with extensive simulations. The proposed mapping algorithm can achieve a mapping error probability below 2%, while the proposed path-planning algorithms extend the experienced radio coverage of the AVs, with limited distance increase with respect to shortest-path existing methods, such as conventional Dijkstra and A* algorithms.

Figures (19)

• Figure 1: Overview of the fog computing scenario.
• Figure 2: Graphical overview of the radio-aware path planning model.
• Figure 3: Generation of the path $\mathbf{C}_M$ from path $\mathbf{W}_K$, and of the map $\mathbf{P}$ from path $\mathbf{C}_M$. Cyan and green pixels represent the synthesized path $\mathbf{C}_M$.
• Figure 4: Expansion of the frontier in the Dijkstra/A* algorithm.
• Figure 5: Localization of the AVs and BSs in the obstacle map, with obstacles and lidar intercept points.