Step and curb detection for autonomous vehicles with an algebraic derivative-based approach applied on laser rangefinder data

Abstract

Abstract-Personal Mobility Vehicles (PMV) is is an important part of the Intelligent Transportation System (ITS) domain. These new transport systems have been designed for urban traffic areas, pedestrian streets, green zones and private parks. In these areas, steps and curbs make the movement of disable or mobility reduced people with PMV, and with standard chair wheels difficult. In this paper, we present a step and curb detection system based on laser sensors. This system is dedicated to vehicles able to cross over steps, for transportation systems, as well as for mobile robots. The system is based on the study of the first derivative of the altitude and highlights the use of a new algebraic derivative method adapted to laser sensor data. The system has been tested on several real scenarios. It provides the distance, altitude and orientation of the steps in front of the vehicle and offers a high level of precision, even with small steps and challenging scenarios such as stairs.

Road Curb Detection: A Historical Survey

“…An analysis of LiDAR2D range measurements in search for road curbs as the closest prominent height step was presented in [26] while the height difference on disparity images to detect curbs on images from a stereo vision system was used in [60]. In [46], a height difference is computed using an estimation of the height difference which is computed using a trapezoidal rule of integration. Height difference on a super voxel grid was used in [56].…”

“…The use of ILP features allows them to reduce the complexity of the algorithm. Other methods that use thresholding on geometric and/or appearance features are reported in [3,4,7,16,17,[24][25][26]36,39,46,48,49,52,71,73].…”

LIDAR-Based road signs detection For Vehicle Localization in an HD Map

“…Processing the whole points cloud is time consuming and is not suitable for real time applications. Therefore, they are usually projected into 2D representations such as 2D orthographic reflectivity grid [13], precise height grid or elevation map [14], [15], or a combination with other information such as colors, curvatures and normals [16]. Finding correspondences between features and map attributes often relies on measuring similarities by defining a specific metric.…”

LIDAR-Based High Reflective Landmarks (HRL)s For Vehicle Localization in an HD Map

“…For lidar sensors, most of state-of-the-art approaches project 3D point clouds to 2D representations such as 2D orthographic reflectivity grid [13], precise height grid or elevation map [14], [15], or a combination with other information such as colors, curvatures and normals [16]. In [13], laser raw data are accumulated to build a reflectivity grid that is matched by directly comparing grid cells to cells from the map.…”