An evolutionary algorithm for online, resource constrained, multi-vehicle sensing mission planning

TL;DR

This paper introduces a genetic algorithm heuristic for online multi-robot sensing mission planning under resource constraints, achieving near-optimal solutions with significantly reduced computation time suitable for real-time applications.

Contribution

It presents a novel GA-based heuristic for the Correlated Team Orienteering Problem, enabling efficient online planning for robotic teams with resource limitations.

Findings

Heuristic solutions are within 5% of optimal performance.

The approach is at least 300 times faster than exact methods.

Execution time is less than a second, suitable for online use.

Abstract

Mobile robotic platforms are an indispensable tool for various scientific and industrial applications. Robots are used to undertake missions whose execution is constrained by various factors, such as the allocated time or their remaining energy. Existing solutions for resource constrained multi-robot sensing mission planning provide optimal plans at a prohibitive computational complexity for online application [1],[2],[3]. A heuristic approach exists for an online, resource constrained sensing mission planning for a single vehicle [4]. This work proposes a Genetic Algorithm (GA) based heuristic for the Correlated Team Orienteering Problem (CTOP) that is used for planning sensing and monitoring missions for robotic teams that operate under resource constraints. The heuristic is compared against optimal Mixed Integer Quadratic Programming (MIQP) solutions. Results show that the quality of the heuristic solution is at the worst case equal to the 5% optimal solution. The heuristic solution proves to be at least 300 times more time efficient in the worst tested case. The GA heuristic execution required in the worst case less than a second making it suitable for online execution.