Abstract
The paper considers the problem of detecting dynamic obstacles on the accumulated occupancymap generated by the computer vision system of a mobile robot. The purpose of this research is toimprove the quality of the obstacle detection algorithm by adding a particle filter to find moving objectsfrom the map data. In the paper, the problem of correct accumulation of data in the occupancymap and reducing the delay in updating the map cells in which the object moves is solved. The modificationof the particle filter presented in the paper is able to work correctly with dynamic obstaclesin a wide range of speeds; it is resistant to outliers caused by random generation of the initial particlesvelocities, and is workable under real conditions in real time in an environment with a lot ofmoving objects. A heuristic has been created that reduces the number of misclassifications in occludedareas. It is shown that the algorithm for detecting dynamic objects in the map is invariant to thetype of sensors used in the vision system, and an implementation combined with an accumulatedoccupancy map is described. The algorithm is implemented and tested on board an autonomous mobilerobot, as well as on an open dataset. The article also provides a comparison with other approachesof dynamic obstacles detection, as well as calculated performance metrics for all analyzedmethods for computers based on the GPU Nvidia RTX 3070 and Jetson AGX Xavier. Promising directionsfor further research to improve the presented algorithm are formulated.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
