松软崎岖地形中基于多物理模型的非完整轮式星球车移动控制研究

The mobility control of wheeled mobile robots that are typical nonholonomic systems is challenging as they don’t satisfy the Brockett’s necessary condition. Fruitful results have been obtained based on the ideal assumptions of “flat terrain”, “no slip” and “no skid”. However, the assumptions do not hold for the planetary rovers that traverse on the deformable and rough terrain, moreover, the rovers are high redundancy driving systems with passive joints, thus many new challengeable issues are brought forward. This project will solve two scientific issues: physics principle of interaction between articulated planetary rovers and the deformable rough terrain, and mobility control of redundantly driving planetary rovers under the condition of coupled slipping and skidding. The coupling interaction principle of terramechanics-kinematics-dynamics of planetary rovers will be revealed, based on which the nominal models and the uncertainty sets are to be derived. A comprehensive multi-time-scale approach of identifying mechanical property and geometry parameters of terrain, as well as the motion state variables will be developed. We will try to make a breakthrough in the design of unified stabilization/tracking robust control law for the nonholonomic planetary rovers when the ideal constraints are broken, and realize the trajectory tracking with minimum energy consumption for redundancy planetary rovers moving on the deformable rough terrain by motion/force hybrid control. We hope to further develop the control theory for the nonholonomic robots through this intensive investigation, and to support the R&D and operation of the field wheeled robots such as the lunar rovers.

轮式机器人作为典型的非完整系统，由于不满足Brockett必要条件而为其移动控制带来了很大挑战，目前形成了基于“平坦地形”和“无滑转”、“无滑移”等理想假设的丰富研究成果。然而，松软崎岖的地形环境导致星球车严重违背了这些假设，且车体是含有多个被动关节的高冗余驱动（6个驱动和4个转向自由度）系统，因此提出了许多挑战性的新问题。本项目拟解决关节式星球车与松软崎岖地形相互作用的物理机制解析、滑转/滑移耦合条件下冗余驱动星球车移动控制两个科学问题，揭示星球车的地面力学-运动学-动力学耦合作用机制，构建多物理标称模型与不确定性集合，建立地形力学/几何参数和运动状态的多时间尺度综合辨识方法，突破理想约束被破坏条件下的非完整星球车镇定/跟踪鲁棒统一控制律设计，实现松软崎岖地形中冗余驱动星球车最小能耗轨迹跟踪的运动/力混合控制,丰富和发展非完整机器人控制理论，为月球车等野外轮式机器人的研发和运行提供支持。

松软崎岖地形环境下的非完整轮式星球车基于轮地力学模型的建立及控制策略研究是星球探测研究中的重要课题。基于轮地力学的星球车基本理论及相关控制策略的研究有利于丰富和发展轮式移动机器人相关理论，为轮式移动机器人建模与控制等的前沿研究提供参考依据。鉴于松软崎岖地形环境下星球车不满足Brockett必要条件，且违背现有研究理论所做的诸多理想假设，本课题针对冗余驱动星球车，研究了轮式移动关节式星球车与松软崎岖地形相互作用的物理机制解析、滑转/滑移耦合条件下冗余驱动星球车移动控制等两个科学问题。揭示了星球车的地面力学-运动学-动力学耦合作用机制，构建了多物理标称模型与不确定性集合，建立了地形力学/几何参数和运动状态的多时间尺度综合辨识方法，实现了理想约束被破坏条件下的非完整星球车镇定/跟踪鲁棒统一控制律设计，完成了松软崎岖地形中冗余驱动星球车最小能耗轨迹跟踪的运动/力混合控制。发表相关学术论文41篇，其中SCI论文28篇，EI论文13篇；获得授权和受理专利12项；获得国家级、省部级和国际奖励4项，其中，“宇航空间机构悬吊法重力补偿关键技术”获得黑龙江省科学技术发明一等奖；会议论文“Switch control for operating constrained mechanisms using a rescuing mobile manipulator with multiple working modes”获评IEEE ARM 2016 会议最佳论文（Best Conference Paper Award）；培养博士、硕士研究生十余人；依托已完成的青年基金面上项目和青年基金面上连续资助项目的研究成果，申请人于2017年获得国际地面车辆系统学会颁发的“ISTVS学会Soehne-Hata-Jurecka Award”，（青年科学家奖，每3年评选1人）。相关成果丰富和发展了非完整轮式星球车建模与控制理论，为星球车等野外轮式机器人的研发提供理论与技术支持。