Research On Mobile Robot's Indoor Navigation Using ROS And Laser Sensor

With the promotion of “The thirteenth five-year plan” robot industry development plan(from 2016 to 2020)and “Made in China 2025”,the rapid development of robot industries has quickly promoted the relevant technology of robotics,including the research of robot navigation.Therefore,research on the key technology of mobile robot's autonomous navigation is of theoretical importance and practical value.Robot Operating System(ROS)is a widely used software platform with good universality and high efficiency.Therefore,based on ROS,two key algorithms: Simultaneous Localization and Mapping(SLAM)algorithm and the path planning algorithm are utilized to design mobile robot's indoor navigation system using laser scanner.Firstly,the SLAM algorithm is studied.For the estimation problem that the RaoBlackwellized particle filter based SLAM(RBPF-SLAM)algorithm used in mobile robots suffers from sample impoverishment in grid mapping,a Gaussian Distributed Resampling based RBPF-SLAM(IRBPF-SLAM)is proposed.Particles are sorted according to their weight size and Gaussian distributed resampling is applied to disperse the high-weight particles so as to generate new particles.Furthermore,by using Gaussian distributed resampling,particle diversity can be maintained and sample impoverishment can be avoided.Thus accurate grid mapping is guaranteed.Experimental results show the effectiveness of the proposed algorithm.Meanwhile,the results prove that the proposed algorithm guarantees reliable estimation with less samples,and the computation burden can be reduced efficiently.Then path planning method is further studied.For the problem of mobile robot's path planning in complex environment with multiple obstacles,an improved artificial bee colony(IABC)based path planning algorithm is proposed.Firstly,a memory mechanism is utilized for the artificial bees to memorize the best paths so the memorized paths can be used to guide the next move in the search progress,improving the search efficiency.Furthermore,maintaining the high fitness policy is applied to avoid best paths from being cast out by the roulette.Experiments show that the improved algorithm guarantees that the mobile robot can safely and quickly finish navigation from the starting point to the destination and it can plan the best path.Finally,ROS based Pioneer 3-DX mobile robot's indoor navigation system using laser scanner is designed and realized.Experiments in different environment demonstrate that the proposed mobile robot navigation system is feasible and reliable.