Research On The Technologies Of Home Service Robot Indoor Autonomous Navigation System

In recent years,with the development of "Industry 4.0" and "Made in China 2025",the home service robot industry has made rapid progress.The autonomous navigation technology is an important symbol of the intelligent home service robots.Therefore,research on home service robot autonomous navigation is of great theoretical significance and practical value.Firstly,the research status of the home service robot navigation technology at home and abroad is expounded in this thesis.On the basis of analyzing the related technologies and common sensors of the home service robot navigation system,the RGB-D,laser ranger finder and odometer are selected as the main sensing devices in the navigation system.And the overall scheme of indoor environments autonomous navigation of the home service robot is designed based on the robot operating system(ROS).In the research of the fast simultaneous location and mapping(FastSLAM)algorithm based on the particle filter,aiming at the problems of low precision and incomplete expression of scanned map information when using a single sensor for FastSLAM,the RGB-D sensor and the laser range finder as the equipment are used to obtain environmental map information,and combines the odometer of the home service robot,a novel fast simultaneous location and mapping based on the multi-sensor information fusion method is proposed.Based on the Bayes probabilistic model,a multi-sensor observation information fusion and selection rule is designed.The unscented Kalman filter is utilized to calculate the importance density fusion in the FastSLAM algorithm,the accuracy of the system can be hold and the running time of the algorithm can be reduced without deducing the Jacobian matrix.The experimental results show that the improved algorithm can construct accurate environmental grid-map.Then,the path planning problem of the service robot is studied and analyzed.Aiming at the problems of convergence slowly and fall into local optimum easily of the whale optimization algorithm,the improved whale optimization algorithm for solving the path planning of the home service robot is put forward in this thesis.In this improved method,a new updating formula of convergence factor is constructed by using the Gauss normal distribution function to balance the global exploration ability and the local development ability of the algorithm.Moreover,the Cauchy mutation factor is introduced in the improved algorithm,the optimal position of the whale individuals are performed mutation.This step can avoid fall into local optimum and premature convergence in the later iteration period of the algorithm.The simulation results show that the improved algorithm has high precision and fast convergence rate,and it can plan the optimal path for the home service robot.Finally,the home service robot indoor autonomous navigation system have been designed and completed according to the FastSLAM algorithm and path planning method mentioned in this thesis.Through the establishment of the home service robot experimental platform,the experiments of grid-map construction and path planning are carried out in real indoor environment.And the final results can show that the home service robot indoor autonomous navigation system is feasible and stable.