Research on Intelligent Vehicle Data Fusion Technology Based on Multiple Sensors

Edge intelligence is the development trend of the integration of Ubiquitous computing and artificial intelligence, and autonomous systems represented by smart cars are playing an increasingly important role in edge intelligence architecture design, integration verification, application services, etc. This article takes the autonomous and accurate indoor mapping of intelligent vehicles as the research object, and systematically designs a boundary point generation scheme that covers exploration, filtering, publishing, and other parts. A hybrid algorithm that integrates global boundary point detection and local boundary point detection is proposed, and a hybrid path planning algorithm based on TEB is proposed for the autonomous exploration problem of intelligent vehicles. Compared to single mapping algorithms such as Gmapping, simulation experiments on the GazeBo and Stage platforms show that the proposed scheme and algorithm have higher accuracy in map fixed-point navigation and can be used as an alternative to traditional human guided autonomous mapping methods.