Hazard detection for small robotic landers and hoppers

Abstract

Planetary hoppers are a new class of vehicle being developed that will provide planetary surface mobility by reusing the landing platform and its actuators to propulsively ascend, translate, and descend to new landing points on the surface of a planetary body. Hoppers enhance regional exploration, with the capability of rapid traverse over 100s-1000s of meters, traverse over hazardous terrain, and exploration of cliffs and craters. These planetary mobility vehicles are fuel limited and as a result, are enabled by carrying sensor payloads that require low mass, low volume, and low on board computational resources. This paper describes a method for hoppers to traverse and land safely in this constrained environment. The method uses optical detection of shadows to provide coarse hazard detection at shallow path angles and long ranges. As a hopper approaches the landing site, a short-range range sensing method is used to determine the local slope and roughness just ahead of the vehicle's flight path. These two techniques are used together to safely land a hopper at downrange locations. The paper presents the method and provides analysis showing the performance and limitations of using this method for hopper traverse.

Simulation Experiment on Landing Site Selection Using a Simple Geometric Approach

“…It happens frequently that the terrain conditions of these area are more complicated, this not only poses a great challenge to the spacecraft Guidance, Navigation, & Control Systems, but also requests more higher demand to the performance of landing obstacle detection and the efficiency of the landing site selection. Up to now, using multiple sensors data for hazard detection have been proposed (Brady et al, 2009;Neveu et al, 2015), including passive optical image data (Bajracharya, 2002;Cheng et al, 2001;Cohanim et al, 2013;Huertas et al, 2006;Mahmood and Saaj, 2015;Matthies et al, 2008;Woicke and Mooij, 2016;Yan et al, 2013), LiDAR data (Amzajerdian et al, 2013;Chakroborty et al, 2009;de Lafontaine et al, 2006;Johnson et al, 2002) and Radar data (Pollard et al, 2003), and so forth.…”