Task Allocation and Trajectory Planning for Swarms Based on Contract Net Protocol and Improved Bee Colony Potential Field Hybrid Algorithm in Adversarial Environments

To address the distributed cooperative combat mission in a typical adversarial environment, a task allocation and trajectory planning algorithm for swarm is proposed, which integrates Contract Net Protocol and Bee Colony Potential Field hybrid method. Dynamic multi-objective task allocation is achieved by the Contract Net Protocol. An improved Bee Colony Potential Field hybrid algorithm is proposed to plan trajectory dynamically for individual agent. Building upon grouping and trajectory information, an improved Rapidly-exploring Random Tree algorithm, combined with Bezier curve, is utilized under a leader-follower model for swarm trajectory optimization. To validate the effectiveness of the proposed algorithm for the red team, the Multi-Agent Deep Deterministic Policy Gradient algorithm is employed to design and train a dynamic hunting strategy model for the blue team. Simulation results demonstrate that the algorithm can efficiently realize intelligent swarm cooperative combat mission and satisfy the requirements of high timeliness and dynamics in adversarial environments.