Autonomous Robot Motion Decision Algorithm Based on Decision Tree Algorithm and Deep Learning Technology

Generally, the traditional robot path planning method can solve the modeling problem well, but the common feature is that there is a lack of search ability in the planning process, which often needs to be combined with other search algorithms. Intelligent method can solve the problem of path planning in complex environment, but some problems will be encountered in the application process. In order to change the problems of traditional robots, such as fixed movements and lack of adaptability, Deep learning (DL) has been gradually introduced into intelligent robot technology. Because the amount of data of environmental parameters that make up the robot is not large, it is easy to obtain a simple data classification rule, and the robot can recognize the environment according to this classification rule. In this paper, an autonomous robot motion decision algorithm based on decision tree and DL technology is proposed, which can complete the functions of robot motion, information collection and processing. Through the overall design of the tracking system, the design of data and robot’s bottom-level instruction transformation and the choice of data transmission mode, a test environment is set up to test the target, and the tracking performance of the improved algorithm is verified.