Redundant Multi-Object Detection for Autonomous Vehicles in Structured Environments

Abstract

This paper presents a redundant multi-object detection method for autonomous driving, exploiting a combination of Light Detection and Ranging (LiDAR) and stereocamera sensors to detect different obstacles. These sensors are used for distinct perception pipelines considering a custom hardware/software architecture deployed on a self-driving electric racing vehicle. Consequently, the creation of a local map with respect to the vehicle position enables development of further local trajectory planning algorithms. The LiDAR-based algorithm exploits segmentation of point clouds for the ground filtering and obstacle detection. The stereocamerabased perception pipeline is based on a Single Shot Detector using a deep learning neural network. The presented algorithm is experimentally validated on the instrumented vehicle during different driving maneuvers.

Validation of scenario-based virtual safety testing using low-cost sensor-based instrumented vehicle

“…Together with vehicle state estimation, the vehicle is able to generate the vehicle trajectory. Feraco et al [24] equipped an instrumented vehicle with a Velodyne LIDAR sensor and a stereo camera for object detection. It will be able to detect multiple objects and map them together in an instant with the computing power of Nvidia Jetson AGX.…”

Automation of haulers for debris removal in tunnel construction

“…In the case of this work, it takes place in a semi‐structured environment, where part of the driving environment is known, but part of the elements does not follow expected rules. Although automated driving is a field that has seen numerous advances in recent years, as is clear from its history, most of the advances are aimed at structured environments (Feraco et al., 2022; Kolski et al., 2006), and, indeed, public works areas are considered semi‐structured areas, ideal for automation purposes, where there are also relevant contributions, in the area of path planning (Dolgov et al., 2010). In addition, it must be considered that public works are not public roads, so the requirements that apply to autonomous road vehicles (especially in relation to regulations) do not apply here.…”