Georgia Institute of Technology
Neuromechanical Control Adaptations in Complex Environments
Abstract
dc:description.abstractAnimals are highly successful locomotors capable of efficient, robust, and adaptive movement across a variety of environments and their effective locomotion strategies have inspired many robotic systems. Despite advancements in technology leading to superior individual robotic components, biological systems can still outperform robots. This gap in performance is attributed to neuromechanical interactions across the many subsystems contained within biological organisms. While studies have looked at these subsystems individually, it remains unclear how effective locomotion strategies emerge from these interactions. This work explores how neuromechanical interactions vary during locomotion in complex environments and how these interactions differ for each distinct appendage. We first show that neuromechanical interactions have a significant impact on the resulting locomotion. We employ two different strategies to do this in robophysical models: 1) we isolate the key components established for insect flight and model asynchronous actuation, a state-dependent actuation strategy to determine the effects of aeromechanical interactions in forward flight on the emergent kinematics, 2) we take a more generalized approach by using a partial information decomposition-based measurement of centralization (strength of neuromechanical interactions between subsystems that respond to perturbations) and independently vary both neural and mechanical interactions to quantify their impact on the resulting control architecture. We then consider limb specific differences in neuromechanical interactions for Blaberus discoidalis : 1) we quantify how the previously mentioned centralization metric varies for the hind and middle limb with increasing terrain complexity to determine if B. discoidalis is capable of varying its neuromechanical control architecture in each limb, 2) we determine when changes in the motor control for the hind and middle limb occur during the transition between two steady states of locomotion, horizontal and inclined running to assess differences in limb neuromechanical responses related to distinct morphology and functions of the limbs. Together these studies demonstrate different methods for quantifying neuromechanical interactions and highlight the importance of the consideration of both neural and mechanical interactions. In addition, the significance of considering these interactions to be distinct between different appendages provides insight to the adaptable neuromechanical architectures of effective biological locomotors.
Degree
thesis:*- Name thesis:degree_name
- Quantitative Biosciences, PhD
- Department thesis:degree_department
- Physics
- Grantor
- Georgia Institute of Technology
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Liu, Ellen
- Advisor dc:contributor.advisor
-
- Sponberg, Simon
- Committee members dc:contributor.committeemember
-
- Goldman, Daniel
- Gravish, Nick
- Chang, Young-Hui
- Sawicki, Gregory
- ORCID person:identifier.orcid
Rights
- Language dc:language.iso
- English
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/82204