Abstract
An integrated system to multi-objective gait planning of bipedal robots is developed. First, a linearisation is defined for kinematic and dynamic models of a 19 degrees of freedom biped. Then the multi-objective gait planning is formulated. It optimises time, power and kinetic energy and singularity avoidance under limitations on actuator torques, workspace, and falling avoidance. An augmented Lagrangian is used to solve the resulting non-linear constrained optimal control problem. To ease difficult calculations of adjoin states; a decoupled form of the biped dynamics is implemented. Simulations are reported showing the effectiveness of the proposed approach.
| Original language | English |
|---|---|
| Pages (from-to) | 96-117 |
| Number of pages | 22 |
| Journal | International Journal of Vehicle Autonomous Systems |
| Volume | 7 |
| Issue number | 1-2 |
| DOIs | |
| State | Published - Aug 2009 |
Keywords
- AL
- Augmented Lagrangian
- Bipedal robots
- Decoupling
- Gradient projection
- Multi-Objective gait planning
- Non-Linear optimal control
ASJC Scopus subject areas
- Control and Systems Engineering
- Automotive Engineering
- Electrical and Electronic Engineering