Research And Implementation Of Autonomous Navigation Robot Platform

The problem of population aging in China has become increasingly prominent and has had a great impact on all aspects of society.Robots play an important role in alleviating the problem of population aging.With the development of artificial intelligence technology,the technical research on intelligent robots continues to deepen.As the basic core of intelligent robot technology,autonomous navigation technology has become a hot spot in scientific research.Many industries have introduced specialized intelligent robots that replace human work,but there is currently no mobile development platform on the market that can collect various functions and enable autonomous navigation testing.Therefore,this paper designs a robot platform with autonomous navigation capability,which can test autonomous navigation algorithm and expand its functions by carrying different devices.The composition location algorithm,path planning algorithm and control algorithm for autonomous navigation are studied.The defects of the algorithm are optimized,improved and tested.The autonomous navigation performance of the robot platform is improved.The specific research and implementation process is as follows:Firstly,according to the performance requirements,the overall framework of autonomous navigation robot platform is designed,including mechanical structure,hardware platform and software system.The mechanical structure is analyzed and designed from three aspects: driving structure,driving mode and hardware specification.According to the performance requirement and mechanical structure,the circuit of hardware platform is analyzed and designed.According to the requirement of autonomous navigation,the structure of driving software and algorithm software in software system is analyzed and designed,so as to complete the framework of autonomous navigation mobile platform.Secondly,the composition localization algorithm for the first half of the selfdirected navigation is studied.The system flow and probability model of SLAM algorithm are sorted out.The advantages and disadvantages of SLAM algorithm based on Kalman filter theory and SLAM algorithm based on particle filter theory are compared and analyzed.On this basis,Fast SLAM algorithm,which combines the advantages of the two algorithms,is deeply studied.A new resampling strategy is introduced to solve the problem of large computation and lack of particles in the algorithm,and the effectiveness of the algorithm is improved by simulation test and analysis.Thirdly,the planning control algorithm of the second half of the autonomous navigation is studied.The map environment and degree of freedom in path planning are determined,and the Dijster-Straw algorithm and A* algorithm in global path planning,artificial potential field method and dynamic window method in local path planning are analyzed and compared.According to the simulation results of the algorithm in matlab,the planning algorithm of the platform is determined,and the defects in the A* algorithm are improved.Finally,according to the real-time requirement of platform control,the PID algorithm is analyzed,and the key parameters are adjusted by simulation software.Fourth,the implementation and test analysis of mobile platform.According to the design framework,the hardware is made,the driver software and algorithm software are programmed,and then the system runs under the ROS framework.After that,the map construction of the actual scene and the performance test of the path planning control algorithm are carried out,and the test results are analyzed and studied.The designed robot platform can carry out autonomous navigation,and the improved composition location algorithm has higher accuracy and shorter path planning length,thus completing the research purpose.