首页 /研究 /Onboard and self-contained landing site selection for planetary landers/hoppers
OTHER

Onboard and self-contained landing site selection for planetary landers/hoppers

Babak E. Cohanim, Jeffrey A. Hoffman, Tye Brady

发表年份
2014
引用次数
3

摘要

This paper describes an operational concept for planetary landing site selection using a long range optical method to do gross site selection, and a short range LIDAR for detailed landing aim point selection. The paper describes the operational concept using the two complementary methods. Simulation and field test results are provided for the proof-of-concept. The first method is a feature-based detection technique method that is developed to rapidly assess relative landing safety. Current hazard detection techniques are very precise, but require sensors that are too large and heavy, and algorithms that require too much computational power for a small robotic explorer. These techniques work under any lighting conditions but are limited at shallow path angles. A new method is needed that can rapidly assess surface hazards and work at shallow path angles. The Size Density Method (SDM) described in this paper enables rapid landing area selection during traverse. This technique is beneficial for small, mass-constrained vehicles, but is agnostic to vehicle size. Even though bigger vehicles may have the capacity to use better sensors, this method will still be applicable. The second method is developed to directly detect and assess the slope and roughness environment at a desired landing location: The Size Density Method proposed in this paper is able to detect rock and crater hazards and provide alternate safe landing area suggestions based on that information but is light-limited and detects hazards coarsely. A complementary algorithm, Sparse Slope and Roughness (SSR), is developed utilizing active sensing techniques to detect hazards at the very end of a traverse, right before landing. This method utilizes short range LIDARs and studies the ability to make accurate hazard detection through sparse sampling of surface features at short range during the final stages of traverse before a vehicle lands. Combined, these two methods provide a powerful operational concept for long range site selection and short range hazard detection to enable safe landing of planetary exploration vehicles.

关键词

TraversePoint cloudComputer scienceRange (aeronautics)LidarTerrainHazardSite selectionSelection (genetic algorithm)Remote sensing

相关论文

查看 OTHER 分类全部论文