ALGORITHM FOR COMPLEXING MULTIPLE DATA SOURCES INTO A SINGLE OCCUPANCY MAP

Abstract

The paper deals with the problem of constructing a passability model of environment with a large number of dynamic objects based on data from several different sensors. The aim of the work is to improve the algorithm for constructing the occupancy map by adding data from both existing algorithms for moving obstacles detection and from millimeter-wave automotive radar. The study solves the problem of combining data on static environment and dynamic objects into one general passability model for further trajectory planning. The modification of the algorithm presented in the article is able to combine data from both occupancy maps based on a threedimensional point cloud from any sensor such as lidar or radar, and arrays of bounding boxes of objects with known coordinates, sizes, and orientation. Data aggregation occurs at the level of building occupancy maps and does not impose requirements on the source of information about dynamic obstacles. The algorithm is able to refine the data on the position and size of the dynamic object by speed from the radar, which allows to plan the trajectory taking into account the movement of dynamic objects. The parallel use of the classical approach allows to detect obstacles in the event of an error in the output of the dynamic obstacle detection algorithm. The developed algorithm works in real time on the Jetson AGX Xavier module, and is tested in real conditions on a mobile robotic platform in autonomous mode. Promising directions for further research to improve the presented approach are formulated.