Research On Robot Navigation Based On Multi-sensor

With the arrival of the fourth wave of the industrial revolution,smart products have entered people’s horizons.Moving robots are one of the important components of robots.Among the key technologies involved in mobile robots,an important issue is robotic navigation.In unknown complex environments,SLAM(Simultaneous Localization and Mapping)technology plays a crucial role in robotics engineering.Since individual sensors can only detect limited information,their application in smart products is limited to sending simple instructions,which can’t real intelligence.At the same time,mobile robots can more accurately perceive the surrounding environment by using multiple sensor technologies.This synergistic working pattern plays a crucial role in mobile robotic navigation and has become a hot spot for current research.(1)When a single sensor is used to build a map,the effect of building a map is not good.By using a variety of sensors and improving the Fast SLAM2.0 particle filter,a new adaptive resampling technology is adopted,which can realize the dynamic adjustment of particles through fuzzy logic,thus obtaining more accurate map results and high-precision maps.(2)The positioning accuracy is poor.The positioning accuracy of mobile robot is improved by optimizing Fast SLAM2.0 algorithm.(3)The existence of the sampling algorithm RRT(Rapidly-Exploiting Random Trees)does not guarantee that the feasible path is relatively optimal.An elliptic sampling method is adopted to replace the global uniform sampling.In order to better meet the sampling requirements of mobile robots,the search strategy of A * algorithm and elliptic search method can realize more efficient result processing and quickly obtain the optimal path.(4)The narrow space makes it difficult for the robot to find the path,which leads to a lot of time spent in the process of finding the path.RRT* algorithm evenly samples the free space,and many redundant branches will be generated in the search tree.The sampling process of RRT*algorithm is improved.Informed-RRT* algorithm is an algorithm obtained by optimizing the sampling process of RRT*.Through carefully designed search process and accurate path planning,navigation efficiency can be improved,and more accurate optimal solution can be obtained,and finally the expected target position can be reached.Finally,a mobile robotic navigation system based on multiple sensors was designed.Through software simulations and practical testing,we found that using this method can deliver more accurate and smoother predictions.Effectively combining the Informed RRT* algorithm with the sampling-based local path planning algoritm to build a better global path,allowing robots to more effectively automate obstacles,improve their safety and efficiency,and experimentally verify the feasibility and effectiveness of the scheme,build maps with an accuracy of about 4 cm,and improve the accurateness of navigation by 0.20432282,hoping to promote the development of robotics technology by providing a reliable basis.