Employing Natural Terrain Semantics in Motion Planning for a Multi-Legged Robot

Recent developments in terrain identification, classification, parameter estimation for the navigation of autonomous robots

“…Since the mapping is prepared globally, the effect of dynamics of the robot will play a key role and hence the application is limited. Belter et al [25] used an intelligent algorithm using RGB-D camera and popular classification algorithms like Support Vector Machine, to identify the terrain and developed a motion planning strategy using A* algorithm. [12] presented an onboard online terrain analyzing mechanism using 3D mapping.…”

Environmental Characterization and Path Planning for Legged Robots Considering Foot-terrain Interaction

“…HOFFMAN 等 [4] 率先开展了面向足式机器人运动规 划的环境建模研究，采用三维激光传感器构建了环 境的栅格地图。BELTER 等 [5] 在此基础上构建了更 灵活的局部地形高程图，使得地图能够跟随机器人 的运动而移动。FANKHAUSER 等 [6] 则提出了一种 概率化的环境建模方法，提高了环境建模方法的可 靠性。此外，部分研究工作开始在环境表征中加入 语义信息，并构建语义地图。提取语义信息的方法 主要有特征法 [7][8] 、词袋法 [9][10] 、基于深度学习的语 义分割法 [11] 等。 基于环境模型进行可通行路径规划，通常先设 置通行评价指标并计算对应的代价地图，然后采用 A*、D*等搜索算法在 2D 代价地图中搜寻代价最低 的目标路径。2009 年，ANNETT 等 [12] 采用地形粗 糙度、倾斜度和阶梯度作为通行评价指标，借助 A* 算法进行六足机器人的全局路径规划，并采用 D* 算法进行局部的路径重规划。 DAVID 等 [13] 基于地形 几何特征进行路径规划并在 BigDog 机器人上得到 成功应用。随着工作环境日趋复杂，高维运动规划 问题引起人们重视，演化出基于采样的规划策略，并 在人形机器人上得到良好实践 [14][15][16] 。最新的研究工作 开始在通行评价指标中引入更丰富的内涵，例如考虑 地形的语义信息进行六足机器人的路径规划 [17]…”

Walking Posture Adaptation for Legged Robot Navigation in Confined Spaces

“…Terrain characterisation for gait adaptation has been shown to improve locomotion efficiency when walking on rough terrain [10]. Elevation maps have been used with characterisation to plan footholds for optimal stability and obstacle avoidance [12], as well as walking over gaps and climbing stairs [11]. Optimising robot dynamics can enable even more dynamic motions [13]; however, none of these methods consider collisions with the body of the robot, which is necessary in confined spaces.…”