Bipedal Robots

Christine Chevallereau is Director of Research at the National Center for Scientific Research (CNRS) and is a researcher at the Institute of Research in Communication and Cybernetics, Nantes, France.Guy Bessonnet is a Professor at the University of Poitiers and is a researcher at the Laboratory of Solid Mechanics at Poitiers, France.Gabriel Abba is Professor of Automatic Control and Robotics at the National Engineer School of Metz and is a researcher at the Laboratory of Industrial Engineering and Mechanical Production, Metz, France.Yannick Aoustin is an Associate Professor at the University of Nantes and is a researcher at the Institute of Research in Communication and Cybernetics, Nantes, France.

• Chapter 1. Bipedal Robots and Walking1.1. Introduction 11.2. Biomechanical approach 21.2.1. Biomechanical system: a source of inspiration 21.2.2. Skeletal structure and musculature 91.3. Human walking 111.3.1. Architecture 111.3.2. Walking and running trajectory data. 131.3.3. Study cases 181.4. Bipedal walking robots: state of the art 211.4.1. A brief history 211.4.2. Japanese studies and creations 241.4.3. The situation in France 271.4.4. General evolution tendencies 311.5. Different applications 321.5.1. Service robotics 331.5.2. Robotics and dangerous terrains 351.5.3. Toy robots and computer animation in cinema 351.5.4. Defense robotics 371.5.5. Medical prostheses 391.5.6. Surveillance robots 401.6. Conclusion 401.7. Bibliography 41Chapter 2. Kinematic and Dynamic Models for Walking2.1. Introduction 472.2. The kinematics of walking 482.2.1. DoF of the locomotion system 482.2.2. Walking patterns 492.2.3. Generalized coordinates for a sagittal step 532.2.4. Generalized coordinates for three-dimensional walking 572.2.5. Transition conditions 662.3. The dynamics of walking 702.3.1. Lagrangian dynamic model 712.3.2. Newton-Euler’s dynamic model 872.3.3. Impact model 982.4. Dynamic constraints. 1032.4.1. CoP and equilibrium constraints 1032.4.2. Non-sliding constraints 1162.5. Complementary feasibility constraints 1172.5.1. Respecting the technological limitations 1182.5.2. Non-collision constraints 1192.6. Conclusion 1232.7. Bibliography 123Chapter 3. Design Tools for Making Bipedal Robots3.1. Introduction 1273.2. Study of influence of robot body masses 1283.2.1. Case 1: the three-link robot 1293.2.2. Case 2: the five-link robot 1473.3. Mechanical design: the architectures carried out 1653.3.1. The structure of planar robots 1653.3.2. 3D robot structures 1683.3.3. Technology of inter-body joints 1723.3.4. Drive technology 1743.4. Actuators 1813.4.1. Actuator types 1813.4.2. Characteristics of electric actuators 1863.4.3. Elements of choice for robotic actuators 1903.4.4. Comparing actuator performances 1933.4.5. Performances of transmission-actuator associations 2023.5. Sensors 2073.5.1. Measuring 2073.5.2. Frequently used sensors 2083.5.3. Characteristics and integration 2093.5.4. Sensors of inertial localization 2103.6. Conclusion 2123.7. Appendix 2133.7.1. Geometric model 2133.7.2. Dynamic model 2133.8. Bibliography 215Chapter 4. Walking Pattern Generators4.1. Introduction 2194.2. Passive and quasi-passive dynamic walking 2204.2.1. Passive walking 2204.2.2. Quasi-passive dynamic walking 2224.3. Static balance walking 2274.4. Dynamic synthesis of walking 2284.4.1. Performance criteria for walking synthesis 2284.4.2. Formalizing the problem of dynamic optimization 2324.5. Walking synthesis via parametric optimization 2364.5.1. Approximating the control variables 2374.5.2. Parameterizing the configuration variables 2384.5.3. Parameterizing the Lagrange multipliers 2464.5.4. Formulation of the parametric optimization problem 2504.5.5. A parametric optimization example 2554.6. Conclusion 2614.7. Bibliography 262Chapter 5. Control5.1. Introduction 2675.2. Hybrid systems and stability study 2695.3. Taking into account the unilateralism of the contact constraint 2735.3.1. Computed torque control 2735.4. Online modification of references 2825.4.1. General principle 2825.4.2. The ZMP’s imposed evolution 2855.4.3. Bounded evolution of the ZMP 2925.5. Taking an under-actuated phase into account 2965.6. Taking the double support phase into account 3015.7. Intuitive and neural network methods 3065.7.1. Intuitive methods 3065.7.2. Neural network method 3115.8. Passive movements 3185.9. Conclusion 3225.10. Bibliography 323Index 327