Research On Indoor Positioning And Path Planning Algorithms

Indoor mobile robots have many advantages such as strong mobility,easy operation and adaptability.With the rapid development of robot technology,they have slowly been integrated into all aspects of our lives,such as home cleaning sweeping robots and goods handling.Warehouse robot.Indoor navigation and positioning and indoor path planning are the key to the realization of indoor mobile robot functions,and also the main content of this paper.This article compares different indoor navigation and positioning algorithms from navigation positioning accuracy and implementation cost,and selects and focuses on indoor Wi Fi positioning.In order to meet the requirements of navigation and positioning in different indoor environments,this paper analyzes a variety of indoor wireless communication channel models as a basis for subsequent indoor Wi Fi positioning research.Indoor Wi FI positioning is mainly divided into two stages: ranging stage and positioning stage.Aiming at the problem of large fluctuations in RSSI received by mobile robots in the ranging stage,the five-point three-time method,the first-average method,and the second-average method are used to process from the perspective of data processing,which effectively reduces RSSI fluctuations.In the positioning stage,five different positioning algorithms were studied,and simulation tests and comparative analysis of different positioning algorithms were carried out.The results show that the best three-point positioning method can achieve positioning accuracy within 2m.In order to further improve the accuracy of indoor navigation and positioning,this paper uses the IMU/Wi Fi combined navigation and positioning solution.Based on Kalman filtering,IMU/Wi Fi loose integrated navigation positioning algorithm and tight integrated navigation positioning algorithm are designed,and simulation experiments show that the integrated navigation positioning algorithm has greatly improved the indoor navigation positioning accuracy.Indoor path planning is another key technology of indoor mobile robots.This paper analyzes and compares different path planning algorithms,focusing on the artificial potential field method,and for the problem of the local minimum value of the artificial potential field method in indoor path planning,an improved algorithm combined with the A-star algorithm is designed.Simulation tests have been carried out in different simulation environments,which shows that the improved algorithm can effectively avoid local minima in indoor path planning.In order to improve the self-learning ability of the mobile robot in the indoor environment,this paper combines the reinforcement learning method in the path planning algorithm.Aiming at the problem that the Q-learning algorithm has multiple convergence times and slow convergence speed,the gravitational potential field is used as the initial value of the Q table,and combined with the environmental trap search,an indoor path planning algorithm based on improved Q-learning is designed.In this paper,by building different indoor simulation environments,the indoor path planning simulation test and comparative analysis of the Q-learning algorithm before and after the improvement are completed.The results show that the improved Q-learning algorithm has a greater number of convergence times and speeds.Promote.Finally,the indoor navigation and positioning algorithm and indoor path planning algorithm studied in this paper are collated and integrated.System comprehensive simulation experiments are carried out in different indoor scenarios to verify the effectiveness and the effectiveness of the indoor navigation and positioning algorithm and indoor path planning algorithm studied.Practicality.