Research On Hybrid Path Planning For Mobile Robots In Indoor Environment

Path planning is one of the core technologies for autonomous navigation of mobile robots,and it ensures that mobile robots can complete tasks efficiently,safely,and autonomously,and has become a research hotspot in robotics.At present,the research on the path planning of mobile robots under the global static environment is relatively mature.However,the research on the path planning of mobile robots under some unknown environment needs to be supplemented and improved.In this thesis,the path planning problem of mobile robot is analyzed based on the indoor environment where the global static environment is known but there are some static or dynamic obstacles that are not detected in advance,and a hybrid path planning method is proposed.For higher-precision environmental maps required for path planning,in this thesis,quantum particle swarm optimization(QPSO)algorithm is introduced into Rao-Blackwillized particle filtering algorithm and a QPSO-RBPF-SLAM algorithm is proposed that in order to solve the traditional Rao-Blackwillized particle filter algorithm's problems of long running time,low precision of the proposed distribution,and particledegeneration in the re-sampling process.On the one hand,this algorithm adds observation information to the basic proposal distribution.On the other hand,it updates the particle pose based on QPSO algorithm during re-sampling,and performs adaptive crossover and mutation operations on high and low weighted particles.Through simulation,it is verified that the algorithm maintains the diversity of particles,effectively mitigates particle degradation,and can obtain higher-precision maps.Next,based on the high-precision environment map obtained,the path planning of the mobile robot is studied.For the globally known static environment,there is shortage of "premature" and the trapping into local optimism in the path planning based on QPSO.An improved QPSO is proposed,that is,an adaptive local search strategy and cross-operation are introduced on the basis of QPSO.Through simulation experiments under different kinds of algorithms,it is verified that the improved QPSO algorithm performance is more superior than other algorithms.For a partially unknown environment,a local path planning method based on the Morphin algorithm is described.With the idea of rolling planning,a rolling window is constructed,and the surrounding environment information is sensed based on sensors carried by the mobile robot itself,and the detected obstacles are identified and analyzed.Meanwhile,The Morphin algorithm is used to avoid static and dynamic obstacles that the mobile robot will collide with.Through simulation and analysis,thismethod can make the mobile robot effectively avoid sudden and static and dynamic obstacles in the unknown environment.This thesis combines the advantages of global path planning and local path planning to propose a hybrid path planning method based on improved QPSO and Morphin algorithm.The improved QPSO is used to plan an optimal global path.The robot moves along the global path.When the robot detects an unknown obstacle,the Morphin algorithm is called to avoid the obstacle and return to the global path to continue moving.The voyager II mobile robot was used to build an experiment platform based on ROS and hybrid path planning experiments were conducted.Through simulation experiments and practical applications,it is verified that the mobile robot can effectively achieve obstacle avoidance and complete path planning on the basis of the global path.