Tham khảo tài liệu 'an introduction to intelligent and autonomous control-chapter 10: learning control systems', công nghệ thông tin, quản trị web phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | 10 Learning Control Systems Jay Farrell and Walter Baker The Charles Stark Draper Laboratory Inc. 555 Technology Square Cambridge MA 02139 Abstract An important capability of an intelligent system is the ability to improve its future performance based on experience within its environment. The concept of learning is usually used to describe the process by which this capability is achieved. The present chapter will focus on control systems that are explicitly designed to have and to exploit this capability. In this context the control system can be viewed as a generally nonlinear mapping from plant outputs to actuation commands so as to achieve certain control objectives with learning as the process of modifying this mapping to improve future closed-loop system performance. The information required for learning that is the information required to correctly generate the desired control system mapping is obtained through direct interactions with the plant and its environment . Thus learning can be used to compensate for a lack of a priori design information by exploiting empirical information that is gained experientially. In this setting the principal benefit of learning control is its ability to accommodate poorly modeled nonlinear dynamical behavior. Contemporary learning control methodologies based on this perspective will be described and compared to traditional control approaches such as gain scheduled robust and adaptive control as well as to earlier learning control paradigms. The discussion and examples that follow will identify both the distinguishing characteristics of learning control systems and the benefits of augmenting traditional control approaches with learning capabilities. 238 INTELLIGENT AND AUTONOMOUS CONTROL SYSTEMS 1. Introduction Control systems for autonomous vehicles must maintain closed-loop system integrity and performance over a wide range of operating conditions. This objective can be difficult to achieve due to a number of circumstances
