Intelligent robot-based material handling vehicles (referred to as intelligent robot) for the transportation of small parts of industrial materials at new energy automobile assembly factories cannot work at the same time in the same interval of compute nodes, which affects the efficiency of material handling. This paper presents a cooperative motion control method for intelligent robots based on a regional distributed structure. Firstly, the motion direction for the intelligent robot is taken as the horizontal axis, a rectangular coordinate system is established, and the robot motion strategy is constructed in the framework of the rectangular coordinate system which was divided into four quadrants. Secondly, the observation space, action space and reward function are constructed based on the Markov decision process. Thirdly, a distributed DQN (Deep Q-Learning) reinforcement learning algorithm is employed to enable decision-making for intelligent robot so that they avoid obstacles and choose the optimal paths. In the context of a MATLAB-based simulation test, the proposed model employs 1 to 20 intelligent robots at the same time, and the obtained material handling success rate is higher than 80%. When 20 to 160 intelligent robots are employed at the same time, the material handling success rate is maintained at 78%. In comparison with the other three employed algorithms, namely the TD3 algorithm, the DQN algorithm and the DDPG algorithm, the convergence speed of the proposed model is the highest and its value can be maintained at about 0.6. By creating an electronic map model of the factory environment and testing the cooperative control of four intelligent robots, the experimental process can be completely carried out. In comparison with the existing tracking model, a reduction of 86 seconds for the accumulated idle time for the intelligent robot indicates that the proposed model can be better applied in real-life contexts.
Intelligent robot, Cooperative control, Observation space, Action space, Reward function, Convergence speed.
Wei LI, Chen CHENG, "Intelligent Robot Cooperative Control Based on Distributed DQN Algorithm", Studies in Informatics and Control, ISSN 1220-1766, vol. 34(1), pp. 57-73, 2025. https://doi.org/10.24846/v34i1y202505
All articles published by Studies in Informatics and Control (SIC) are published in accordance with the terms and conditions of the Creative Commons Attribution License (CC BY-NC-SA 4.0).