Abstract
In the near future, the communication between autonomous cars will produce a network of sensors that will allow us to know the state of the roads in real time. Lidar technology, upon which most autonomous cars are based, allows the acquisition of 3D geometric information of the environment. The objective of this work is to use point clouds acquired by Mobile Laser Scanning (MLS) to segment the main elements of road environment (road surface, ditches, guardrails, fences, embankments, and borders) through the use of PointNet. Previously, the point cloud was automatically divided into sections in order for semantic segmentation to be scalable to different case studies, regardless of their shape or length. An overall accuracy of 92.5% has been obtained, but with large variations between classes. Elements with a greater number of points have been segmented more effectively than the other elements. In comparison with other point-by-point extraction and ANN-based classification techniques, the same success rates have been obtained for road surfaces and fences, and better results have been obtained for guardrails. Semantic segmentation with PointNet is suitable when segmenting the scene as a whole, however, if certain classes have more interest, there are other alternatives that do not need a high training cost.
Highlights
IntroductionMore complex sensors have been added to capture the environment, such as cameras and LiDAR sensors [1]
As technology evolves, cars integrate a number of sensors that make the automation of simple processes associated with driving possible, such as a brightness sensor for automatic lighting, a rain sensor for activating windscreen wipers, and a temperature sensor for a climate control system.Recently, more complex sensors have been added to capture the environment, such as cameras and LiDAR sensors [1]
The input data is a point cloud of a road environment acquired with Mobile Laser Scanning (MLS) and the trajectory followed by the system during the acquisition
Summary
More complex sensors have been added to capture the environment, such as cameras and LiDAR sensors [1]. These sensors are essential for Advanced Driver Assistance Systems (ADAS). Autonomous vehicles must have knowledge at all times of its immediate environment in order to decide and interact with it. Almost all autonomous vehicles base their perception of the built environment on LiDAR technology. LiDAR technology makes it possible to acquire the near environment of the vehicle quickly, accurately, in 3D, in true magnitude, and regardless of the light conditions. With LiDAR data, vehicles can automatically detect pedestrians [2], other vehicles [3], road signs [4], and curbs [5,6]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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
