Autonomous planetary exploration using LIDAR data

Abstract

Abstract-In this paper we present the approach for autonomous planetary exploration developed at the Canadian Space Agency. The goal of this work is to autonomously navigate to remote locations, well beyond the sensing horizon of the rover, with minimal interaction with a human operator. We employ LIDAR range sensors due to their accuracy, long range and robustness in the harsh lighting conditions of space. Irregular Triangular Meshes (ITMs) are used for representing the environment providing an accurate yet compact spatial representation. In this paper a novel path-planning technique through the ITM is introduced, which guides the rover through flatter terrain and safely away from obstacles. Experiments performed in CSA's Mars emulation terrain that validate our approach are also presented. I. INTRODUCTIONMobile robotics has enabled scientific breakthroughs in planetary exploration [1]. Recent accomplishments have demonstrated beyond doubt the necessity and feasibility of semi-autonomous rovers for conducting scientific exploration on other planets. Both Mars Exploration Rovers (MERs) "Spirit" and "Opportunity" have the ability to detect and avoid obstacles, picking a path that would take them along a safe trajectory. MER's have reached traverses of 300m/sol. On occasion, the rovers have had to travel to locations that were at the fringe of the horizon of their sensors or even slightly beyond.The next rover missions to Mars are the "Mars Science Laboratory" (MSL) [2] and ESA's ExoMars [3]. Both of these missions have set target traverse distances on the order of one kilometer per day. Both the MSL and ExoMars rovers are therefore expected to drive regularly a significant distance beyond the horizon of their environment sensors. Earthbased operators will therefore not know a-priori the detailed geometry of the environment and will thus not be able to select via-points for the rovers throughout their traverses.One of the key technologies that will be required is the ability to sense and model the 3D environment in which the rover has to navigate. To address the above mentioned issues, the Canadian Space Agency is developing a suite of technologies for long-range rover navigation. For the purposes of this paper, "long-range" is defined as a traverse that takes the rover beyond the horizon of the rover's environment sensors.In the next Section we discuss the state-of-the-art in robotic planetary exploration. Section II presents the overall process for planetary exploration together with a short description of our test-bed. Next we present a summary of our approach to environmental modelling, Section IV

Training a terrain traversability classifier for a planetary rover through simulation

“…18 Neural networks would seem worthy of investigation for the three-dimensional (3D) LiDAR classification problem. Work has been done by other researchers on classifying aerial LiDAR data that generate elevation maps of limited resolution, 19,20 and some significant work has been done in the case of 3D LiDAR from a rover perspective 4,9,10,[21][22][23] and more specifically other types of classifiers such as Markov random fields, 24 Bayesian classifiers, 25 support vector machines (SVMs) 26 and fuzzy modelling. 11 Hata et al successfully use a neural network to do such classification.…”

Low Resolution Lidar-Based Multi-Object Tracking for Driving Applications

“…Optical cameras may fail perceiving correctly the environment in certain conditions, therefore other robust sensors are needed. Lidar technology provides the desired robustness and precision under harsh conditions [21], which make laser scanners suit for autonomous driving applications. Although lidar sensors are expensive compared to artificial vision technologies, recently more affordable models are being introduced into the market.…”

3D registration of laser range scenes by coincidence of coarse binary cubes

“…These sensors have also been proposed for scene perception and modeling in mobile robot applications such as simultaneous localization and mapping [5], planetary exploration [25], and surveillance [2].…”