Efficient Model Identification for Tensegrity Locomotion

This paper aims to identify in a practical manner unknown physical parameters, such as mechanical models of actuated robot links, which are critical in dynamical robotic tasks. Key features include the use of an off-the-shelf physics engine and the Bayesian optimization framework. The task being considered is locomotion with a high-dimensional, compliant Tensegrity robot. A key insight, in this case, is the need to project the space of models into an appropriate lower dimensional space for time efficiency. Comparisons with alternatives indicate that the proposed method can identify the parameters more accurately within the given time budget, which also results in more precise locomotion control.

Authors

• Shaojun Zhu
• David Allen Surovik
• Kostas E. Bekris
• Abdeslam Boularias

Keywords

• Robots
• Engines
• Optimization
• Physics
• Predictive models
• Dimensionality reduction
• Task analysis
• Model Identification
• Dynamical
• Actuator
• Low-dimensional Space
• Bayesian Optimization
• Robotic Tasks
• Physics Engine
• Training Data
• Complex Systems
• Dynamic Model
• Parameter Space
• Dynamic Network
• Equations Of Motion
• Complex Dynamics
• Latent Space
• Optimal Policy
• Lowest Error
• Original Space
• Markov Decision Process
• Variational Autoencoder
• Original Parameters
• Toy Example
• Simple Dynamic Model
• Two-layer Neural Network
• Random Search
• Empirical Rules
• Policy Search
• Model-based Approach
• Input State