Research On Indoor Integrated Navigation Of Mobile Robot Based On ROS

With the continuous innovation of artificial intelligence mobile robot technology,the improvement of social living standard and the continuous improvement of indoor facilities,robots gradually integrate into all aspects of our daily life.And the demand and requirements for indoor navigation are increasing.Different navigation systems have their own characteristics and disadvantages.At present,single and traditional navigation can not meet the basic requirements of indoor positioning.The integration of various navigation methods can optimize the system,but the hardware level requires a high platform.The calculation amount of the algorithm on the software end becomes larger.This paper is devoted to the design of mobile robot integrated navigation system for indoor environment.After investigation,a combination system of lidar,inertial navigation and hall encoder is selected.The research focuses on multi-sensor fusion,slam technology,path path planning,etc.And then provides theoretical basis and decision support for indoor integrated navigation of mobile robot.The main contents include:(1)The theory of indoor integrated navigation is described,including coordinate system conversion,radar working principle,inertial navigation principle and robot motion control.In view of the indoor needs of mobile robots to consider different scenes,the surrounding environment needs to be scanned and built on the premise of autonomous navigation.The radar data is analyzed by ICP scan matching algorithm.Considering the accumulated error of single sensor,the information of inertial sensor and odometer sensor is used as the data source of mobile robot position and attitude calculation.(2)In order to meet the needs of stable navigation in indoor environment,this paper estimates and modifies the parameters and noise of the model based on adaptive extended Kalman filter to reduce the actual error caused by the filter.The effectiveness of the algorithm is shown by simulation experiments,and the performance is improved by about 20%.Path planning is the main navigation technology of indoor mobile robot.This paper analyzes and compares different classical path planning algorithms,and focuses on a* algorithm.In view of the problems of a* algorithm in indoor path planning,the evaluation function,node search and optimal path check of a* algorithm are modified.The simulation results show that the time and distance of path planning are shortened by the algorithm improvement,the number of turns is reduced by 44%,and the turning degree is reduced by 50%.(3)In order to further verify the effectiveness of integrated navigation,a hardware integration and software platform based on ROS(robot operating system)mobile robot is built.Three different scenes are set up,namely indoor simple environment,single obstacle environment and complex obstacle environment,which can complete the autonomous navigation obstacle avoidance and environment plan construction of mobile robot.The experimental results show that the fusion composition distortion degree of multi-source sensor fusion is small,and the navigation distance and time of improved a* algorithm path planning are optimized.The results show that the integrated navigation system designed in this paper can realize the basic functions,and the stability of the operation is improved.