Real-Time LIDAR-Based Urban Road and Sidewalk Detection for Autonomous Vehicles.

Ernő Horváth,Claudiu Pozna,Miklós Unger

doi:10.3390/s22010194

Ernő Horváth, Claudiu Pozna + Show 1 more

Open Access

https://doi.org/10.3390/s22010194

Copy DOI

Abstract

Road and sidewalk detection in urban scenarios is a challenging task because of the road imperfections and high sensor data bandwidth. Traditional free space and ground filter algorithms are not sensitive enough for small height differences. Camera-based or sensor-fusion solutions are widely used to classify drivable road from sidewalk or pavement. A LIDAR sensor contains all the necessary information from which the feature extraction can be done. Therefore, this paper focuses on LIDAR-based feature extraction. For road and sidewalk detection, the current paper presents a real-time (20 Hz+) solution. This solution can also be used for local path planning. Sidewalk edge detection is the combination of three algorithms working parallelly. To validate the result, the de facto standard benchmark dataset, KITTI, was used alongside our measurements. The data and the source code to reproduce the results are shared publicly on our GitHub repository.

Highlights

Urban Road and Sidewalk DetectionAutonomous vehicles, known as self-driving vehicles, are currently being rapidly improved
Sensing is basically data acquisition from the sensors; perception means feature extraction from the sensor data; planning means to create a feasible trajectory; control is responsible for executing this trajectory
The sidewalk edge detection is the combination of the introduced star-shaped search, X-zero, and Z-zero methods in 3D voxels

Summary

Introduction

Autonomous vehicles, known as self-driving vehicles, are currently being rapidly improved. These vehicles or robots have the capability of sensing their environment and moving safely without human input. Sensing is basically data acquisition from the sensors; perception means feature extraction from the sensor data; planning means to create a feasible trajectory; control is responsible for executing this trajectory. The scope of this work is perception, more precisely the extraction of road and sidewalk features. This information will be provided to the planning subsystem. Our focus is on LIDAR, which is a common light ranging sensor

Objectives

Results

Conclusion