The presented book aims to describe possible frameworks for setting up nonlinear design problems that have to be solved in the context of robots trying to understand, interact with, and manipulate their environments. Robots are highly nonlinear systems and even though they can be linearized under some restrictive assumptions, most practical scenarios require the design of nonlinear controllers to work around uncertainty and measurement-related issues. It turned out over the years that Lyapunov’s direct method is an extremely effective tool to both design and analyze controllers for robotic systems. Chapter 1 begins by providing a brief history of robotics. It is followed by an introduction to the Lyapunov-based design philosophy. The chapter ends with a practical note by describing the evolution of real-time control design systems and the associated operating environments and hardware platforms that they are based upon. In Chapter 2 a quick introduction to a host of standard control design tools available for robotic systems is provided. In order to prepare for the chapters ahead, all these techniques are analyzed in a common Lyapunov-based framework. Computed torque methods, where the model nonlinearities are canceled through exact model knowledge, and the system robustness to unmodeled disturbances is discussed under PD control, continuous robust control, and sliding mode control. Adaptive control techniques are discussed when the model is uncertain. The chapter closes with discussing the challenges of designing control laws for redundant link robot manipulators when the control objectives are stated in the task-space of the robot. Chapter 3 discusses some problems in visual servoing control. The first problem addressed is that of a robot end-effector tracking a prerecorded time-varying reference trajectory under visual feedback from a monocular camera. The next problem addressed in this chapter is that of estimating the shape of a continuum robot. Traditional position sensing devices such as encoders cannot be used in this situation since it is not easy to define links and joints for such robots. The third problem dealt with in this chapter is that of designing homography-based visual servo control methods for solving tracking and regulation robots in the context of wheeled mobile robots. The final problem addressed in Chapter 3 is the classic Structure From Motion (SFM) problem, specifically the development of an adaptive nonlinear estimator in order to identify the Euclidean coordinates of future points on a object based upon relative movement between that object and a single fixed camera. Chapter 4 deals with the problems of path planning and control for manipulator arms and wheeled mobile robots, both when the obstacle locations are known a-priori and when they need to be determined in real time using fixed or in-hand vision as a active feedback element. The first problem addressed is path following using Velocity Field Control (VFC) - this technique can be applied when it is more critical to follow a contour exactly than it is to track a desired time-varying trajectory. Another application of path-following is when a Navigation Function (NF) approach is utilized to create a path around obstacles to a desired goal location. As an extension, it is shown how VFC- and NF-based techniques can be utilized to solve the obstacle avoidance problem for mobile robots. The final portion of this chapter deals with the design of an image space extremum seeking path planner such that in a singularity free PBVC controller can be designed that works on visual feedback and is able to reject lens distortion and uncertainties in the camera calibration. The Chapter 5 deals with the emerging research area of human-machine interaction. The chapter begins by exploring smart exercise machines that provide optimal physical training for the user by altering the machine’s resistance based on user performance. Steer-by-wire control of vehicles is discussed next, with the focus being on locking the steering response of the vehicle to the user input as well as ensuring that the road feel experienced by the user can be appropriately adjusted. The third problem addressed in this chapter is that of teleoperator systems where the focus is on both facilitating the application of desired motion and desired force in the remote environment by the user, as well as ensuring that the system is safety able to reflect desired forces back to the user. The final topic addressed in Chapter 5 is that of rehabilitations robot which is safety able to direct user limb motions along selectable trajectories in space that optimize their rehabilitation after disease or injury. The book is aimed at graduate students and researchers who would like to understand the application of Lyapunov-based control design techniques to emerging problems in robotics. It assumes a background in undergraduate level linear control theory. Some knowledge on nonlinear systems and Lyapunov-based design techniques for such systems may be desirable.
Reviewer: Bojidar Cheshankov (Sofia)