Analytical and Experimental Investigation of Force Controlled Off-Road Wheeled Vehicles

本研究计画系探讨非道路轮式车辆之主动控制。有效控制这些系统中的力分布，特别是控制所有车轮-地面接触力，对于实现这些车辆的运动和悬架自由度的可接受控制是至关重要的。在不平坦的地形上，轮式车辆具有运动学和准静态力分布特性，这些特性与车辆在平坦地形上所具有的特性有着根本的不同。如果车辆-地形组合被瞬时视为具有闭合运动学回路的空间联动，则平坦地形导致这些虚拟空间联动中的固有对称性。这种对称性导致这些车辆的运动学和静力分布行为的奇异性。文献中的传统车辆甚至表面操作不解决这些奇异性，因为这些车辆名义上研究的平面运动机器的小扰动，在其悬挂自由度。运动和悬挂通常被视为解耦问题。在不平坦的地形上，对称性和奇点并不总是存在的。使用基于螺旋系统理论的运动学几何框架，开发了对轮式车辆在不平坦地形上的移动性（自由度的数量）和运动潜力（可用运动）的理解。这个框架也可以用来研究这些车辆的力分布特性，通过扩展率运动学/静力分析的二元性，已经观察到在机器人的经典问题，这些种空间机构。运动学几何框架特别适合于研究主动越野车辆，因为它们的系统设计通常不包括有意的被动顺应性（集中或分布式）。扩展的运动学几何方法的刚体动力学的复杂系统，称为空间矢量力学是用来研究这些越野车辆的动态行为。该项目包括“分析/数字”和“实验”两个方面。分析/数值目标包括一个空间矢量力学（SVM）框架，结合螺旋系统理论和牛顿-欧拉动力学使用空间矢量符号的发展。支持向量机框架允许在一个综合的方式进行调查;开发通用的主动越野轮式车辆的运动学合成技术;开发一个模块化的动态模拟器进行数值研究的两类轮式车辆;和开发系统级的力控制策略，主动越野车辆。实验任务包括越野实验的实验原型称为力控三脚架（FCT）。FCT是一类主动车辆的代表，其包括可以主动改变其轴长的两轮轴。

This research project deals with an active control of offroad wheeled vehicles. An effective control of the force distribution in these systems, particularly the control of all the wheel-terrain contact forces, is central to achieving acceptable control of the motion and suspension degrees of freedom of these vehicles. On uneven terrain, wheeled vehicles possess kinematic and quasi-static force distribution characteristics that are fundamentally different from those that vehicles possess on even terrain. If the vehicle-terrain combination are instantaneously treated as a spatial linkage with closed kinematic loops, an even terrain leads to an inherent symmetry in these virtual spatial linkeage. This symmetry leads to singularities in kinematic and static force distribution behavior of these vehicles. The literature on conventional vehicles for even surface operation does not address these singularities since these vehicles are nominally studied as planar motion machines with small perturbations in their suspension degrees of freedom. Motion and suspension are generally treated as decoupled problems. On uneven terrain, the symmetry and hence the singularities are not always present. Using a kinematic-geometric framework based on screw system theory, an understanding of the mobility (number of degrees of freedom) and the motion potential (available motion) of wheeled vehicles on uneven terrain is developed. This framework can also be used to study the force distribution characteristics of these vehicles by extending the rate kinematics/static force analysis duality that has been observed in classical problems in robotics to these kinds of spatial mechanisms. The kinematic-geometric framework is particularly suited for studying active off-road vehicles since their system designs typically do not include intentional passive compliances (lumped or distributed). An extension of the kinematic-geometric approach to rigid body dynamics of complex systems, known as spatial vector mechanics is used to study the dynamic behavior of these offroad vehicles. The project includes both the `analytical/numerical' and `experimental' aspects. The analytical/ numerical objectives include the development of a Spatial Vector Mechanics (SVM) framework that combines Screw System Theory and Newrton-Euler Dynamics using Spatial Vector Notation. The SVM framework allows the investigation in an integrated manner; to develop general kinematic synthesis techniques for active off-road wheeled vehicles; to develop a modular dynamic simulator for performing numerical studies of two classes of wheeled vehicles; and to develop system-level force control strategies for active off road vehicles. The experimental task include off-road experiments with an experimental prototype known as the Force controlled Tricycle (FCT). The FCT is representative of a class of active vehicles that include two-wheel axles that can actively vary their axle lengths.