Mapping with an Autonomous Car

Abstract

In this paper, we present an approach towards mapping and safe navigation in real, large-scale environments with an autonomous car. The goal is to enable the car to autonomously navigate on roads while avoiding obstacles and while simultaneously learning an accurate three-dimensional model of the environment. To achieve these goals, we apply probabilistic state estimation techniques, network-based pose optimization, and a sensor-based traversability analysis approach. In order to achieve fast map learning, our system compresses the sensor data using multi-level surface maps. The overall systems runs on a modified Smart car equipped with different types of sensors. We present several results obtained from extensive experiments which illustrate the capabilities of our vehicle.