Research On Sound Source Location And Path Planning Technology Based On SLAM

With the continuous improvement of robot intelligence,robots have been widely used in various fields such as home life,intelligent warehousing,factory workshops,medical guidance,etc.Therefore,higher requirements have been put forward for robot perception technology.And robot auditory perception provides a more natural,convenient,and efficient way of communication for human-computer interaction,and its related research has been highly concerned.In recent years,sound source localization technology based on microphone arrays has provided effective solutions for robot auditory perception.However,there are still problems such as low accuracy and poor real-time performance of sound source localization.Therefore,this article explores an improved algorithm for sound source localization for mobile robots,and combines it with robot displacement measurement to achieve the mapping of sound source localization results to SLAM maps,and completes the path planning from the robot to the sound source target.This study provides theoretical reference and technical reserves for sound source localization and tracking of mobile robots,and has certain practical value.The main research content is summarized as follows:An improved SRP-PHAT-HSDA sound source localization algorithm has been proposed.This method adopts a hierarchical search strategy,which sequentially scans the space using coarse resolution and fine resolution grids to reduce the number of memory searches;And using a microphone directionality model to reduce the number of scanning directions,ignoring unimportant microphone pairs,further reducing computational load;At the same time,the MSW automatic calibration mechanism has been added,effectively improving the real-time positioning accuracy while ensuring the accuracy of sound source positioning.For sound source tracking,this paper proposes an improved three-dimensional Kalman(M3K)method that can simultaneously track the direction of sound sources in three-dimensional space.Completed sound source localization experiments based on the ODAS platform,verifying the effectiveness and superiority of the algorithm.Based on laboratory environmental mapping testing,four classic 2D laser SLAM algorithms were compared and analyzed,and the advantages and disadvantages of each algorithm were summarized.The Gmapping algorithm was selected to establish an environmental map,laying the foundation for sound source localization and path planning.A sound source localization method based on robot displacement measurement is proposed to address the issue of the inability to estimate the distance between sound sources in far-field sound source localization.The sound source coordinates are obtained by combining robot displacement and sound source direction angle calculation.Through coordinate conversion,the mapping of sound source positioning results to SLAM maps is achieved.A corresponding improved algorithm has been proposed to address the security issues of traditional A* path planning algorithms.By adopting the method of diagonal non passing,adding judgment conditions and filtering sub nodes,the reliability of obstacle avoidance is improved to a certain extent,and the path planning of the robot to the sound source target is completed.The feasibility and effectiveness of the proposed algorithm were verified through simulation experiments.We have built a mobile robot sound source localization experimental platform and conducted relevant experimental research.Using a four wheeled intelligent robot equipped with a microphone array as the hardware platform and ROS system as the software platform,sound source angle measurement experiments,sound source localization experiments,and path planning experiments from the robot to the sound source target were conducted in a laboratory environment.The effectiveness of the improved sound source localization algorithm proposed in this paper and the reliability of the improved A* path planning algorithm were verified.