This chapter describes the control principles necessary for an articulated biped model to accomplish balanced locomotion during walking and climbing. We explain the synthesizes mechanism for coordinated control of lower-body joints (i.e., ankle, hip, and knee). A humanoid biped can have a large number of degrees of freedom (DOF) that make it challenging to create physically correct, plausible and efficient motions. While we are able to define the physical principles of unintelligent models (e.g., multi-rigid body systems), the area of actively controlling a virtual character to mimic real-world creatures is an ongoing area of research. We focus on the control strategy and stability factors during continuous motion for the performing of essential rudimentary tasks (i.e., walking and climbing). We use a multi-level feedback mechanism to generated motion trajectories for the different actions, such as, stepping and walking. For example, the support leg is controlled through active forces (i.e., actuated joint feedback) based upon the control strategy to create a targeted set of parabolic trajectories for the action (e.g., stepping or climbing). The parabolic trajectories control the articulated skeleton while taking into account environmental influences (e.g., terrain height and balance information); with control parameters, such as leg-length, centre-of-mass (COM) location, and step-length being fed-back into the control mechanism.
|Title of host publication||Dynamic Balancing of Mechanisms and Synthesizing of Parallel Robots|
|Editors||Dan Zhang, Bin Wei|
|Number of pages||10|
|Publication status||Published - 2016|
- Control architecture
- Trajectory generation
ASJC Scopus subject areas