Conference proceeding
Robotics Application of a Method for Analytically Computing Infinitesimal Phase Response Curves
Biomimetic and Biohybrid Systems, pp.104-115
Lecture Notes in Computer Science
12/23/2020
DOI: 10.1007/978-3-030-64313-3_12
Abstract
This work explores a method for analytically computing the infinitesimal phase response curves (iPRCs) of a synthetic nervous system (SNS) for a hybrid exoskeleton. Phase changes, in response to perturbations, revealed by the iPRCs, could assist in tuning the strength and locations of sensory pathways. We model the SNS exoskeleton controller in a reduced form using a state-space representation that interfaces neural and motor dynamics. The neural dynamics are modeled after non-spiking neurons configured as a central pattern generator (CPG), while the motor dynamics model a power unit for the hip joint of the exoskeleton. Within the dynamics are piecewise functions and hard boundaries (i.e. “sliding conditions”), which cause discontinuities in the vector field at their boundaries. The analytical methods for computing the iPRCs used in this work apply the adjoint equation method with jump conditions that are able to account for these discontinuities. To show the accuracy and speed provided by these methods, we compare the analytical and brute-force solutions.
Details
- Title: Subtitle
- Robotics Application of a Method for Analytically Computing Infinitesimal Phase Response Curves
- Creators
- Marshaun N FitzpatrickYangyang Wang - University of Iowa, MathematicsPeter J ThomasRoger D QuinnNicholas S Szczecinski
- Resource Type
- Conference proceeding
- Publication Details
- Biomimetic and Biohybrid Systems, pp.104-115
- Publisher
- Springer International Publishing; Cham
- Series
- Lecture Notes in Computer Science
- DOI
- 10.1007/978-3-030-64313-3_12
- ISSN
- 0302-9743
- eISSN
- 1611-3349
- Language
- English
- Date published
- 12/23/2020
- Academic Unit
- Mathematics; Iowa Neuroscience Institute
- Record Identifier
- 9984084400102771
Metrics
8 Record Views