A Market-Based 3D-Aware Coordination Framework for Multi-Robot Systems in Dynamic Environments

Abstract

Efficient task allocation and coordination are critical challenges in multi-robot systems, particularly in dynamic and unstructured environments. This study proposes a market-based coordination framework that leverages local spatial awareness and decentralized decision-making to optimize task distribution, reduce energy consumption, and minimize collisions among autonomous robots. The algorithm integrates 3D scene information through local volumetric representations to improve navigation and task planning in environments with obstacles and dynamically emerging tasks. Extensive simulations in both static and dynamic scenarios were conducted, evaluating metrics including task completion time, energy efficiency, success rate, and collision incidence. Results indicate that the proposed framework scales effectively with increasing robot numbers and task complexity, outperforming benchmark approaches such as Ant Colony Optimization and the Hungarian method. Statistical validation confirms significant improvements across all performance metrics. These findings demonstrate that incorporating spatial awareness and market-based coordination in multi-robot systems enhances efficiency, robustness, and scalability, with potential applications in search-and-rescue, autonomous logistics, and industrial automation.