Dual-Arm Robot Trajectory Planning Based on Deep Reinforcement Learning under Complex Environment

Abstract

In this article, the trajectory planning of the two manipulators of the dual-arm robot is studied to approach the patient in a complex environment with deep reinforcement learning algorithms. The shape of the human body and bed is complex which may lead to the collision between the human and the robot. Because the sparse reward the robot obtains from the environment may not support the robot to accomplish the task, a neural network is trained to control the manipulators of the robot to prepare to hold the patient up by using a proximal policy optimization algorithm with a continuous reward function. Firstly, considering the realistic scene, the 3D simulation environment is built to conduct the research. Secondly, inspired by the idea of the artificial potential field, a new reward and punishment function was proposed to help the robot obtain enough rewards to explore the environment. The function is consisting of four parts which include the reward guidance function, collision detection, obstacle avoidance function, and time function. Where the reward guidance function is used to guide the robot to approach the targets to hold the patient, the collision detection and obstacle avoidance function are complementary to each other and are used to avoid obstacles, and the time function is used to reduce the number of training episode. Finally, after the robot is trained to reach the targets, the training results are analyzed. Compared with the DDPG algorithm, the PPO algorithm reduces about 4 million steps for training to converge. Moreover, compared with the other reward and punishment functions, the function used in this paper will obtain many more rewards at the same training time. Apart from that, it will take much less time to converge, and the episode length will be shorter; so, the advantage of the algorithm used in this paper is verified.

Keywords: complex environment; deep reinforcement learning; dual-arm robot; reward; trajectory planning.

Figure 1

The graphical abstract of this article.

Figure 2

The overall structure of the robot.

Figure 3

The agent–environment interaction in a Markov decision process.

Figure 4

The relationship and structure diagram of the three major components of ML agents.

Figure 5

Training logic diagram.

Figure 6

The target setting diagram.

Figure 7

Training scheme.

Figure 8

Collision detection configuration diagram. (a) Robot collision detection configuration. (b) Human body and bed collision detection configuration diagram.

Figure 8

Collision detection configuration diagram. (a) Robot collision detection configuration. (b) Human body and bed collision detection configuration diagram.

Figure 9

Diagram of Barrier point.

Figure 10

PPO algorithm network training flowchart.

Figure 11

Training environment.

Figure 12

The Agent training process of generating trajectory.

Figure 13

Diagram of training results. The number in the upper right corner of the picture represents the moving order of the robot.

Figure 14

Training results diagram. (a) Cumulative Reward. (b) Episode Length.

Figure 15

The human posture changing diagram.

Figure 16

The training results in posture changing.

Figure 17

The training result of DDPG.