Mobie Robot Localization And Map Building In The Dynamic Environment

Robot map building in the dynamical and complex environment has been a hot issue in roboticsresearch domain. When there are dynamic obstacles in the environment, laser sensor data forscanning will have an error match and meanwhile duo to the limitations of sensor robot it willgenerate a global inconsistent map and lead to robot global localization failure in the dynamical andcomplex environment. To address the issues, this paper proposes a dynamic obstacle detection andfilter dummy data points’ method, the global optimization algorithm of fusing stochastic gradientdescent method and the nonlinear least squares, sample-based matching rules extend Monte Carloalgorithm. The main innovations are as follows:(1) First, the dynamic obstacle detection and filter dummy data points’ approach is proposed.Taking out three consecutive observations from the laser sensor scanning data points and combininggiven error function formula and cumulative function formula to determine whether the obstaclesare dynamic obstacles or not. When the obstacle is a dynamical obstacle, the paper by filtering thedata points to build grid map.(2) Secondly, due to the limitations of robot’s sensor and the noise in the environment, it willbring the question of global map inconsistency. This paper presents fusing stochastic gradientdescent method and nonlinear least squares optimization algorithm to simplify the computationalcomplexity and to avoid the problem of high side iterative. After the optimization, the local map canobtain the global maximum likelihood map.(3) Finally, to achieve global localization for mobile robots in a dynamic and complexenvironment, this paper presents sample-based matching rules extend Monte Carlo algorithm. Theextend Monte Carlo algorithm is based on the traditional Monte Carlo positioning algorithm, butwhen the particle set is poor during the sampling, this paper changes the traditional Monte Carloresampling method to fusing perception information and motion information for dimensionalresampling. The algorithm can effectively reduce particles’ poverty and improve particles’ diversityto achieve mobile robot global localization in the dynamic and complex environment.