基于多智能体深度强化学习的多星观测任务分配方法OA北大核心CSTPCD
Multi-Satellite Observation Task Allocation Method Based on Multi-Agent Deep Reinforcement Learning
为应对多星环境中复杂多约束条件下的任务分配场景,提出一种多星自主决策观测任务分配算法,该算法采用基于集中式训练、分布式执行的多智能体深度强化学习算法.通过这种方式训练后的卫星智能体,即使在没有中心决策节点或通信受限的情况下,仍具有一定的自主协同能力及独立实现多星观测任务的高效分配能力.
To address the task allocation scenario under complex and constrained conditions in a multi-satellite environment,a multi-satellite autonomous decision-making observation task allocation algorithm is proposed.The algorithm uses a multi-agent deep reinforcement learning algorithm based on centralized training and distributed execution.The satellite agents trained by this algorithm have certain autonomous collaboration capabilities and the ability …查看全部>>
王桢朗;何慧群;周军;金云飞
上海卫星工程研究所,上海 201109上海航天技术研究院,上海 201109上海卫星工程研究所,上海 201109上海卫星工程研究所,上海 201109
计算机与自动化
多智能体系统深度强化学习多星系统多智能体深度确定性策略梯度算法任务规划
multi-agent systemdeep reinforcement learningmulti-satellite systemmulti-agent deep deterministic policy gradient(MADDPG)mission planning
《上海航天(中英文)》 2024 (1)
108-115,8
评论