摘要
机器人控制中的计算资源需求量越来越大,移动边缘计算技术是一种解决该问题的手段;针对现在研究中的算法无法满足服务器多机器人协同任务场景中的求解需求,设计了一种简化策略算法,首先将问题简化为单服务器多机器人问题,而后使用强化学习算法进行求解,最后通过仿真实验与随机卸载及本地卸载进行对比,验证了算法的有效性。
The demand for computing resources in robot control is increasing, and mobile edge computing technology is a means to solve this problem. In view of the fact that the algorithm under study cannot meet the solution requirements in the server multi-robot collaborative task scenario, a simplified method is designed.The strategy algorithm firstly simplifies the problem into a single-server multi-robot problem, and then uses the reinforcement learning algorithm to solve it. Finally, the simulation experiment is compared with random unloading and local unloading to verify the effectiveness of the algorithm.
作者
李少波
崔好
白洪飞
刘意杨
Li Shaobo;Cui Hao;Bai Hongfei;Liu Yiyang(Key Laboratory of Networked Control Systems,Chinese Academy of Sciences,Shenyang 110016;Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 110016;Institutes for Robotics and Intelligent Manufacturing,Chinese Academy of Sciences,Shenyang 110169;University of Chinese Academy of Sciences,Beijing 100049)
出处
《仪器仪表标准化与计量》
2022年第3期15-17,共3页
Instrument Standardization & Metrology
关键词
移动边缘计算
强化学习
协同计算
计算卸载
Mobile Edge Computing
Reinforcement Learning
Collaborative Computing
Compute Unload
