Chen J., Jia B., Zhang K. Multi-View Geometry Based Visual Perception and Control of Robotic Systems

CRC Press, 2018. — 361 p. — ISBN: 0815365985.This book describes visual perception and control methods for robotic systems that need to interact with the environment. Multiple view geometry is utilized to extract low-dimensional geometric information from abundant and high-dimensional image information, making it convenient to develop general solutions for robot perception and control tasks. In this book, multiple view geometry is used for geometric modeling and scaled pose estimation. Then Lyapunov methods are applied to design stabilizing control laws in the presence of model uncertainties and multiple constraints.
Over the past decade, there has been a rapid development in the vision-based perception and control of robotic systems. Especially, multiple-view geometry is utilized to extract low-dimensional geometric information from abundant and high-dimensional image space, making it convenient to develop general solutions for robot perception and control tasks. This book aims to describe possible frameworks for setting up visual perception and control problems that need to be solved in the context of robotic systems.
The visual perception of robots provides necessary feedback for control systems, such as robot pose information, object motion information, and drivable road information. Since 3D information is lost and image noise exists in the imaging process, the effective pose estimation and motion identification of objects are still challenging. Besides, mobile robots are generally faced with complex scenes making it difficult to robustly detect drivable road space for safe operation. In this book, multiple-view geometry is exploited to describe the scene structure and maps from image space to Euclidean space. Optimization and estimation theories are applied to reconstruct the geometric information of the scene.
Then, it is convenient to identify the real-time states of the robot and objects and to detect drivable road region based on geometric information.Foundations.
Robotics.
Multiple-View Geometry.
Vision-Based Robotic Systems.
Visual perception of robotics.
Introduction to Visual Perception.
Road Scene 3D Reconstruction.
Recursive Road Detection with Shadows.
Range Identification of Moving Objects.
Motion Estimation of Moving Objects.
Visual control of robotics.
Introduction to Visual Control.
Visual Tracking Control of General Robotic Systems.
Robust Moving Object Tracking Control.
Visual Control with Field-of-View Constraints.
Visual Control of Mobile Robots.
Trifocal Tensor Based Visual Control of Mobile Robots.
Unified Visual Control of Mobile Robots with Euclidean Reconstruction.
Appendices.