Adaptive Neural Control of Walking Robots

Citation

Randall, M.J. (2002), "Adaptive Neural Control of Walking Robots", Industrial Robot, Vol. 29 No. 5, pp. 469-469. https://doi.org/10.1108/ir.2002.29.5.469.1

Adaptive Neural Control of Walking Robots establishes a theoretical framework for the control structure for an autonomous walking robot. The research presents a neural network for controlling the joints of a hexapod robot which is capable of exploring a rough‐terrain environment with sparse footholds.

The book comprises ten chapters which address two main focal areas. After an Introductory chapter, chapters 2 to 5 establish the theoretical basis for the control of walking robots. Chapter 2, A Generic Intelligent Control Hierarchy, addresses the advantages of a hierarchical approach, a definition of “autonomy” for mobile machines, and related work. A justification for biologically inspired engineering, the anatomy and leg structure of insects, insect behaviours, and rough‐terrain strategies, are amongst the topics discussed in chapter 3. Chapters 4 and 5 discuss Models for Co‐ordination of Walking, and Leg Trajectory Planning and Generation, respectively.

The remainder of the book focuses on the stable on‐line control of complex structures, namely a hexapod based on the stick insect Carausius morasus. Chapter 6 presents Hexapod Kinematics and Dynamics, and addresses topics including forward and inverse kinematics; inverse dynamics; force distribution using an optimisation criterion; and comparative dynamics in legged systems. The Theory of Stable Adaptive Neural Control for Open‐Chain Systems is discussed in chapter 7, while chapter 8 addresses Stable Neural Control Systems with Constraints and Closed Kinematic Chains.

Leg design, body design, controller hardware and software, adaptive neural control during walking, rough‐terrain adaptation, and basic gait experiments are discussed in chapter 9, Hexapod Experiments. The remaining chapter of the book provides a summary, conclusions and possible avenues for future research.

This is a very well written reference text which will be of interest to students, academics and professional engineers alike. Most chapters include high level coverage of the topics making them accessible to non‐specialists. Conversely, the original contributions of the research make it relevant to the professional engineering community.