Global Path Planning of Lunar Rover Under Static and Dynamic Constraints
- PDF / 13,367,472 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 36 Downloads / 240 Views
ORIGINAL PAPER
Global Path Planning of Lunar Rover Under Static and Dynamic Constraints Ji Hoon Bai1 · Young-Jae Oh1 Received: 30 April 2019 / Revised: 4 September 2019 / Accepted: 21 February 2020 © The Korean Society for Aeronautical & Space Sciences 2020
Abstract This paper proposes a global path planning for lunar rovers in polar regions under static and dynamic constraints. Prior to lunar mission launch, a specific mission path must be set for a lunar rover. The mission path is generated under uncertain information. Only lunar digital elevation model (DEM) and secondary parameters derived from DEM are available. The parameters can be divided into two categories: static parameters and dynamic parameters. Parameters such as DEM, slope, and roughness can be categorized into static parameters. Illumination and thermal inertia can be categorized into dynamic constraints. This paper introduces such parameters as constraints for path planning, by either giving weight to each parameters or suggesting a threshold for a dead zone. A* path planning method is used to implement such parameters as constraints. Simulations of path planning are shown. Keywords Path planning · Static and dynamic parameters · Digital elevation model · Lunar rover
1 Introduction Since the age of renaissance, mankind’s curiosity regarding outer space has grown. From Galileo Galilei to Stephen Hawking, many astronomers have devoted themselves to understand what lies beyond Earth. Back then, the only way to perceive outer planets was to use a telescope. Since then, technology has grown ever faster and mankind has finally been able to launch space vehicles such as satellites and planetary rovers. However, since the cost of launching a planetary rover is doubtlessly high, system redundancy must be satisfied for unexpected outcomes and the mission path must be optimized for maximum exploration output. Due to the importance of the mission, many scientists and engineers have dedicated themselves to the area of mission path planning. Path planning or motion planning can be categorized into several research topics: discrete planning, samplebased planning, combinatorial planning, feedback planning, decision-theoretic planning, and differential con-
B 1
Young-Jae Oh [email protected] Department of Aerospace Engineering, Korea Advanced Institute of Science and Technology, Daejeon 34141, South Korea
straints planning [1]. Discrete planning is the simplest algorithm among the listed methods. A*, Dijkstra’s algorithm, forward/backward search, and bidirectional search are well-known algorithms for discrete planning. Samplebased motion planning consists of random sampling, rapidly exploring random trees (RRTs), probabilistic roadmaps, and many more. Combinatorial motion planning includes Canny’s algorithm, maximum-clearance roadmaps, and Davenport–Schinzel sequences. Feedback motion planning consists of navigation functions, smooth manifolds, numerical potential functions, etc. Decision-theoretic planning contains Bayesian classifications, zero-sum games th
Data Loading...
