Research Of Mobile Robot Indoor Navigation Technology Based On Multi-Sensors Information Fusion

In recent years,artificial intelligence and robot technology have developed rapidly.At the same time,with the coming of information age,information fusion technology have become an important technology.Therefore,it is very important for mobile robot indoor navigation based on information fusion.Therefore,on the platform of the robot operating system,the scheme of mobile robot indoor navigation system based on information fusion is completed by using the odometer,the inertial measurement unit and the laser sensor.In the study of FastSLAM algorithm for mobile robots,a modified FastSLAM algorithm based on information fusion is proposed to solve the problem of low precision mapping using a single sensor for FastSLAM.The EKF information fusion algorithm is used to integrate the odom data with the inertial measurement unit data to provide more accurate data for the proposed distribution.The improved QPSO(Quantum-behaved Particle Swarm Optimization)algorithm is used in the particle resampling.In the particle updating formula of the QPSO algorithm,the dynamic threshold based on particle fitness is introduced,and the corresponding particles are divided according to the threshold.The corresponding variation can increase particle diversity and reduce the possibility of particle degradation.Experimental results show that the improved FastSLAM algorithm is feasible and effective.In order to solve the problem of path planning of the mobile robot in a more complex indoor environment,a path planning method based on global planning combined with local planning is proposed.In the global path planning,the improved QPSO algorithm and Dijstra algorithm are used.Firstly,a suboptimal path is planned by Dijstra algorithm,and then the suboptimal path is optimized by the improved QPSO algorithm.The improved QPSO algorithm introduces improved clustering factor in the compression expansion factor to determine the division of the search phase,and introduces an improved staged mutation strategy in the particle update position formula to effectively prevent rapid reduction of the population diversity of the particle swarm.Experimental result shows that robots can plan shorter paths more quickly.In the local path planning,using the improved APF path planning,the fuzzy controller in the fuzzy control algorithm adjusts the gravitational coefficient in the gravitational formula and improves the repulsive coefficient in the repulsive force formula.The experimental results show that the robot can get a smoother and shorter path.In the path planning method based on global planning combined with local planning,a global short path is planned for the robot using the global path planning method.The improved artificial potential field algorithm is used to smooth the path on the planned path.The experimental results show that the robot can quickly plan an optimal path.Finally,the design and implementation of mobile robot indoor navigation system based on information fusion on the Pioneer 3-DX mobile robot platform is completed.The mapping and path planning are completed through experiments.The results show that the indoor navigation system of mobile robot based on multi-sensor information fusion is feasible and effective.