松软地形下移动机器人的运动和规划研究

松软地形下机器人的运动和规划研究是移动机器人领域的研究热点。移动机器人在松软地形上运动的过程中伴随着滑移和沉陷，这两种现象严重地降低了机器人的移动性能。近年来该领域的研究主要分为机器人的结构设计，轮壤接触的力学建模以及路径规划等方向。这些研究尝试将机器人运动过程中的沉陷和滑移参数化，并用于机器人控制规划任务中。目前的研究多集中于以星球车为代表的轮式机器人领域，但是轮式机器人的参数化方法难以适用于其它运动模态的移动机器人，例如足式机器人。为解决这些问题，本文在多模态机器人平台上进行了以下研究：
（1） 建立多模态机器人在松软斜坡上的综合模型，并利用该模型估计机器人在松软斜坡上的滑移参数和力学性能。通过对轮壤接触力的定量分析，本文提出了移动机器人在松软地形上的姿态控制策略和轮壤接触优化策略。
（2） 提出了一种适用于多模态机器人与松软地形的接触参数定义。该定义将接触过程划分为两个接触阶段和三个状态点，并利用相机等传感器测量运动过程中的这三个状态点和对应的接触信号。接下来将接触信息输入 CNN 网络用以估计接触参数。该参数估计方法能够准确的估计多种交互状态下的接触参数，准确
率达到 96% 以上。
（3） 设计了基于机器人运动不确定性的路径规划算法。该方法考虑了轮足复合机器人在松软地形上的运动不确定性，并给出了对应的计算方法。基于接触参数数据集，本文给出了具体的运动不确定性参数，并将这些参数应用于路径规划任务中。仿真结果表明，该路径规划算法能够降低所生成路径近 15% 的运动不
确定性。

The motion and planning of robots over soft terrain are the diffculties in the field of mobile robot research. When a mobile robot moves on soft terrain, there are two common phenomena, slipping and sinking, which seriously restrict the mobility of the robot. In recent years, researchers in this field have mainly focused on these areas to improve robot mobility, such as robot design. These studies attempt to parameterize the sinkage and slippage and use them in robot control or planning tasks. However, most of the current research focuses on the field of wheeled vehicles, and these methods are diffcult to apply to multi-modal mobile robots, such as walking robots. In this paper, the following studies are carried out through a hybrid wheel-footed robot platform:
（ 1） This thesis builds an integrated model of a multi-modal robot on soft slopes and uses this model to estimate the robot’s contact parameters and mechanical properties. Based on the analysis of the wheel-soil contact force, this paper provides an attitude control strategy and the wheel-soil contact optimization strategy of the mobile robot on soft terrain.
（ 2） This thesis proposes a definition of contact parameters suitable for hybrid wheel-foot robots and soft terrain. Based on the definition, the contact parameters during the interaction process between hybrid robots and soft terrain are estimated. After processing the contact signal during the interaction process, this thesis inputs the contact information into the CNN to estimate these parameters. The CNN-based method can accurately estimate the contact parameters in various interaction states, and the accuracy rate can reach more than 96%.
（ 3） A path planning algorithm based on robot motion uncertainty is designed. This method considers the robot motion uncertainty in the soft terrain and gives the corresponding calculation method of the motion uncertainty. The motion uncertainty parameters are applied to the path planning task. The simulation results show that the path planning algorithm can effectively reduce the motion uncertainty of the planned path (more than 15%).
 

[1] MCKEE M. Mars rover may not escape sand trap for weeks[J]. New Scientist, 2009: 2-4.
[2] LINDEMANN R A, VOORHEES C J. Mars exploration rover mobility assembly design, test and performance[C]//2005 IEEE International Conference on Systems, Man and Cybernetics:volume 1. IEEE, 2005: 450-455.
[3] PATEL N, SLADE R, CLEMMET J. The exomars rover locomotion subsystem[J]. Journal of Terramechanics, 2010, 47(4): 227-242.
[4] WONG J Y. Theory of ground vehicles[M]. John Wiley & Sons, 2008: 117-142.
[5] ISHIGAMI G. Terramechanics-based analysis and control for lunar/planetary exploration robots [J]. PhD Thesis, Graduate School of Engineering, Tohoku University, 2008: 40-74.
[6] IAGNEMMA K, SHIBLY H, DUBOWSKY S. On-line terrain parameter estimation for planetary rovers[C]//Proceedings 2002 IEEE international conference on robotics and automation: volume 3. IEEE, 2002: 3142-3147.
[7] JOHNSON J B, KULCHITSKY A V, DUVOY P, et al. Discrete element method simulations of mars exploration rover wheel high-slip mobility tests[M]//Earth and Space 2014. 2014: 183-191.
[8] JIA Z, SMITH W, PENG H. Fast computation of wheel-soil interactions for safe and effcient operation of mobile robots[C]//2011 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2011: 3004-3010.
[9] JIA Z, SMITH W, PENG H. Terramechanics-based wheel–terrain interaction model and its applications to off-road wheeled mobile robots[J]. Robotica, 2012, 30(3): 491-503.
[10] CUNNINGHAM C, ONO M, NESNAS I, et al. Locally-adaptive slip prediction for planetary rovers using gaussian processes[C]//2017 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2017: 5487-5494.
[11] REINA G, MILELLA A, PANELLA F. Vision-based wheel sinkage estimation for rough-terrain mobile robots[C]//2008 15th International Conference on Mechatronics and Machine Vision in Practice. IEEE, 2008: 75-80.
[12] DING L, GAO H, DENG Z, et al. Slip-ratio-coordinated control of planetary exploration robots traversing over deformable rough terrain[C]//2010 IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE, 2010: 4958-4963.
[13] LIU Z, GUO J, DING L, et al. Online estimation of terrain parameters and resistance force based on equivalent sinkage for planetary rovers in longitudinal skid[J]. Mechanical systems and signal processing, 2019, 119: 39-54.
[14] GAO H, GUO J, DING L, et al. Longitudinal skid model for wheels of planetary exploration rovers based on terramechanics[J]. Journal of Terramechanics, 2013, 50(5-6): 327-343.
[15] DING L, GAO H, DENG Z, et al. Foot–terrain interaction mechanics for legged robots: Modeling and experimental validation[J]. The International Journal of Robotics Research, 2013, 32 (13): 1585-1606.
[16] XIONG X, AMES A D, GOLDMAN D I. A stability region criterion for flat-footed bipedal walking on deformable granular terrain[C]//2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2017: 4552-4559.
[17] GOSYNE J R, HUBICKI C M, XIONG X, et al. Bipedial locomotion up sandy slopes: Systematic experiments using zero moment point methods[C]//2018 IEEE-RAS 18th International Conference on Humanoid Robots (Humanoids). IEEE, 2018: 994-1001.
[18] HUBICKI C M, AGUILAR J J, GOLDMAN D I, et al. Tractable terrain-aware motion planning on granular media: an impulsive jumping study[C]//2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2016: 3887-3892.
[19] CHANG A H, HUBICKI C M, AGUILAR J J, et al. Learning terrain dynamics: A gaussian process modeling and optimal control adaptation framework applied to robotic jumping[J]. IEEE Transactions on Control Systems Technology, 2020, 29(4): 1581-1596.
[20] RIESER J, QIAN F, GOLDMAN D. Legged-locomotion on inclined granular media[C]//APS March Meeting Abstracts: volume 2016. 2016: Y40-003.
[21] ARSLAN Ö, SARANLI U. Reactive planning and control of planar spring–mass running on rough terrain[J]. IEEE Transactions on Robotics, 2011, 28(3): 567-579.
[22] KOLVENBACH H, BÄRTSCHI C, WELLHAUSEN L, et al. Haptic inspection of planetary soils with legged robots[J]. IEEE Robotics and Automation Letters, 2019, 4(2): 1626-1632.
[23] KOLVENBACH H, BELLICOSO D, JENELTEN F, et al. Effcient gait selection forquadrupedal robots on the moon and mars[C]//14th International Symposium on Artificial Intelligence, Robotics and Automation in Space (i-SAIRAS 2018). 2018: 1-8.
[24] ARM P, ZENKL R, BARTON P, et al. Spacebok: A dynamic legged robot for space exploration [C]//2019 international conference on robotics and automation. IEEE, 2019: 6288-6294.
[25] INOTSUME H, CREAGER C, WETTERGREEN D, et al. Finding routes for effcient and successful slope ascent for exploration rovers[C]//The International Symposium on Artificial Intelligence, Robotics and Automation in Space (i-SAIRAS). 2016: 1-8.
[26] INOTSUME H, SUTOH M, NAGAOKA K, et al. Modeling, analysis, and control of an actively reconfigurable planetary rover for traversing slopes covered with loose soil[J]. Journal of Field Robotics, 2013, 30(6): 875-896.
[27] SIM B S, KIM K J, YU K H. Development of body rotational wheeled robot and its verification of effectiveness[C]//2020 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2020: 10405-10411.
[28] KOLVENBACH H, ARM P, HAMPP E, et al. Traversing steep and granular martian analog slopes with a dynamic quadrupedal robot[J]. Journal of Field Robotics, 2021: 01-30.
[29] SKONIECZNY K, MORELAND S J, ASNANI V M, et al. Visualizing and analyzing machinesoil interactions using computer vision[J]. Journal of Field Robotics, 2014, 31(5): 820-836.
[30] ROTHROCK B, KENNEDY R, CUNNINGHAM C, et al. Spoc: Deep learning-based terrain classification for mars rover missions[M]//AIAA SPACE 2016. 2016: 5539.
[31] IWASHITA Y, NAKASHIMA K, RAFOL S, et al. Mu-net: Deep learning-based thermal ir image estimation from rgb image[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2019: 0-0.
[32] ZHOU R, DING L, GAO H, et al. Mapping for planetary rovers from terramechanics perspective [C]//2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2019: 1869-1874.
[33] MIZUNO M, KUBOTA T. A new path planning architecture to consider motion uncertainty in natural environment[C]//2020 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2020: 2182-2188.
[34] MACDONALD R A, SMITH S L. Active sensing for motion planning in uncertain environments via mutual information policies[J]. The International Journal of Robotics Research, 2019, 38 (2-3): 146-161.
[35] TSANG F, WALKER T, MACDONALD R A, et al. Lamp: Learning a motion policy to repeatedly navigate in an uncertain environment[J]. IEEE Transactions on Robotics, 2021: 1-15.
[36] FANKHAUSER P, HUTTER M. A universal grid map library: Implementation and use case for rough terrain navigation[M]//Robot Operating System (ROS). Springer, 2016: 99-120.
[37] HAUSER K. Semi-infinite programming for trajectory optimization with non-convex obstacles [J]. The International Journal of Robotics Research, 2021, 40(10-11): 1106-1122.
[38] VARAVA A, CARVALHO J F, KRAGIC D, et al. Free space of rigid objects: Caging, path non-existence, and narrow passage detection[J]. The international journal of robotics research, 2021, 40(10-11): 1049-1067.
[39] DEPTULA P, LICITRA R A, ROSENFELD J A, et al. Online approximate optimal path-planner in the presence of mobile avoidance regions[C]//2018 Annual American Control Conference (ACC). 2018: 2515-2520.
[40] QIAN F, KODITSCHEK D E. An obstacle disturbance selection framework: emergent robot steady states under repeated collisions[J]. The International Journal of Robotics Research, 2020, 39(13): 1549-1566.
[41] CHIANG H T, MALONE N, LESSER K, et al. Path-guided artificial potential fields with stochastic reachable sets for motion planning in highly dynamic environments[C]//2015 IEEE international conference on robotics and automation (ICRA). IEEE, 2015: 2347-2354.
[42] ELHOSENY M, THARWAT A, HASSANIEN A E. Bezier curve based path planning in a dynamic field using modified genetic algorithm[J]. Journal of Computational Science, 2018, 25: 339-350.
[43] CIMURS R, HWANG J, SUH I H. Bezier curve-based smoothing for path planner with curvature constraint[C]//2017 First IEEE International Conference on Robotic Computing (IRC). 2017: 241-248.
[44] PRÁGR M, ČÍŽEK P, FAIGL J. Traversal cost modeling based on motion characterization for multi-legged walking robots[C]//2019 European Conference on Mobile Robots (ECMR). IEEE, 2019: 1-6.
[45] INOTSUME H, SUTOH M, NAGAOKA K, et al. Evaluation of the reconfiguration effects of planetary rovers on their lateral traversing of sandy slopes[C]//2012 IEEE International Conference on Robotics and Automation. IEEE, 2012: 3413-3418.
[46] YANG C, DING L, TANG D, et al. Analysis of the normal bearing capacity of the terrain in case of foot-terrain interaction based on terzaghi theory[C]//2016 IEEE International Conference on Robotics and Biomimetics. IEEE, 2016: 443-448.
[47] KARAFIATH L L, EA N. Soil mechanics for off-road vehicle engineering.[M]. 1978: 3-8.
[48] ANDRADE G, AMAR F, BIDAUD P, et al. Modeling robot-soil interaction for planetary rover motion control[C]//Proceedings. 1998 IEEE/RSJ International Conference on Intelligent Robots and Systems: volume 1. 1998: 576-581.
[49] INOTSUME H, KUBOTA T, WETTERGREEN D. Robust path planning for slope traversing under uncertainty in slip prediction[J]. IEEE Robotics and Automation Letters, 2020, 5(2): 3390-3397.
[50] PIPERAKIS S, KOSKINOPOULOU M, TRAHANIAS P. Nonlinear state estimation for humanoid robot walking[J]. IEEE Robotics and Automation Letters, 2018, 3(4): 3347-3354.
[51] INOTSUME H, SUTOH M, NAGAOKA K, et al. Evaluation of the reconfiguration effects of planetary rovers on their lateral traversing of sandy slopes[C]//2012 IEEE International Conference on Robotics and Automation. 2012: 3413-3418.
[52] VAN DEN BERG J, ABBEEL P, GOLDBERG K. Lqg-mp: Optimized path planning forrobots with motion uncertainty and imperfect state information[J]. The International Journal of Robotics Research, 2011, 30(7): 895-913.
[53] BRY A, ROY N. Rapidly-exploring random belief trees for motion planning under uncertainty [C]//2011 IEEE international conference on robotics and automation. IEEE, 2011: 723-730.
[54] MELCHIOR N A, KWAK J Y, SIMMONS R. Particle rrt for path planning in very rough terrain [C]//NASA Science Technology Conference 2007 (NSTC 2007). Citeseer, 2007.
[55] MELCHIOR N A, SIMMONS R G, et al. Particle rrt for path planning with uncertainty.[C]// ICRA. Citeseer, 2007: 1617-1624.
[56] MIZUNO M, KUBOTA T. A new path planning architecture to consider motion uncertainty in natural environment[C]//2020 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2020: 2182-2188.
[57] ANGELOVA A, MATTHIES L, HELMICK D, et al. Learning and prediction of slip from visual information[J]. Journal of Field Robotics, 2007, 24(3): 205-231.
[58] INOTSUME H, KUBOTA T, WETTERGREEN D. Robust path planning for slope traversing under uncertainty in slip prediction[J]. IEEE Robotics and Automation Letters, 2020, 5(2): 3390-3397.