Mammalian locomotion is generated by central pattern generators (CPGs) in the spinal cord, which produce alternating flexor and extensor activities controlling the locomotor movements of each limb. Afferent feedback signals from the limbs are integrated by the CPGs to provide adaptive control of locomotion. Responses of CPG-generated neural activity to afferent feedback stimulation have been previously studied during fictive locomotion in immobilized cats. Yet, locomotion in awake, behaving animals involves dynamic interactions between central neuronal circuits, afferent feedback, musculoskeletal system, and environment. To study these complex interactions, we developed a model simulating interactions between a half-center CPG and the musculoskeletal system of a cat hindlimb. Then, we analyzed the role of afferent feedback in the locomotor adaptation from a dynamic viewpoint using the methods of dynamical systems theory and nullcline analysis. Our model reproduced limb movements during regular cat walking as well as adaptive changes of these movements when the foot steps into a hole. The model generates important insights into the mechanism for adaptive locomotion resulting from dynamic interactions between the CPG-based neural circuits, the musculoskeletal system, and the environment.
more »
« less
An Adaptive Frequency Central Pattern Generator for Synthetic Nervous Systems
Generators (CPGs) are being used for controlling the complicated gaits and timing
required for stable walking. Traditionally, these models are precisely designed
for oscillation at a set of specific frequencies and phase relationships, which while
easier to design is not conducive to robust and stable walking.
In recent years, work has been done on designing adaptive models of CPGs.
These CPGs are able to exhibit complex behaviors such as learning the resonant
dynamics of a system to improve walking stability, as well as using mathematical
learning rules to learn arbitrary signals and embed their relationships
within the system.
This work explores the possibility of implementing an adaptive frequency
CPG with a similar behavior to these systems, using conductance-based models
of dynamic non-spiking neurons connected as a synthetic nervous system
(SNS).
more »
« less
- Award ID(s):
- 1704436
- NSF-PAR ID:
- 10076340
- Date Published:
- Journal Name:
- Lecture notes in computer science
- Volume:
- 10928
- ISSN:
- 1611-3349
- Page Range / eLocation ID:
- 361-364
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Key points The neuromotor system generates flexible motor patterns that can adapt to changes in our bodies or environment and also take advantage of assistance provided by the environment.
We ask how energy minimization influences adaptive learning during human locomotion to improve economy when walking on a split‐belt treadmill. We use a model‐based approach to predict how people should adjust their walking pattern to take advantage of the assistance provided by the treadmill, and we validate these predictions empirically.
We show that adaptation to a split‐belt treadmill can be explained as a process by which people reduce step length asymmetry to take advantage of the work performed by the treadmill to reduce metabolic cost.
Our results also have implications for the evaluation of devices designed to reduce effort during walking, as locomotor adaptation may serve as a model approach to understand how people learn to take advantage of external assistance.
Abstract In everyday tasks such as walking and running, we often exploit the work performed by external sources to reduce effort. Recent research has focused on designing assistive devices capable of performing mechanical work to reduce the work performed by muscles and improve walking function. The success of these devices relies on the user learning to take advantage of this external assistance. Although adaptation is central to this process, the study of adaptation is often done using approaches that seem to have little in common with the use of external assistance. We show in 16 young, healthy participants that a common approach for studying adaptation, split‐belt treadmill walking, can be understood from a perspective in which people learn to take advantage of mechanical work performed by the treadmill. Initially, during split‐belt walking, people step further forward on the slow belt than the fast belt which we measure as a negative step length asymmetry, but this asymmetry is reduced with practice. We demonstrate that reductions in asymmetry allow people to extract positive work from the treadmill, reduce the positive work performed by the legs, and reduce metabolic cost. We also show that walking with positive step length asymmetries, defined by longer steps on the fast belt, minimizes metabolic cost, and people choose this pattern after guided experience of a wide range of asymmetries. Our results suggest that split‐belt adaptation can be interpreted as a process by which people learn to take advantage of mechanical work performed by an external device to improve economy.
-
Sampling-based motion planning algorithms provide a means to adapt the behaviors of autonomous robots to changing or unknown a priori environmental conditions. However, as the size of the space over which a sampling-based approach needs to search is increased (perhaps due to considering robots with many degree of freedom) the computational limits necessary for real-time operation are quickly exceeded. To address this issue, this paper presents a novel sampling-based approach to locomotion planning for highly-articulated robots wherein the parameters associated with a class of locomotive behaviors (e.g., inter-leg coordination, stride length, etc.) are inferred in real-time using a sample-efficient algorithm. More specifically, this work presents a data-based approach wherein offline-learned optimal behaviors, represented using central pattern generators (CPGs), are used to train a class of probabilistic graphical models (PGMs). The trained PGMs are then used to inform a sampling distribution of inferred walking gaits for legged hexapod robots. Simulated as well as hardware results are presented to demonstrate the successful application of the online inference algorithm.more » « less
-
Sampling-based motion planning algorithms provide a means to adapt the behaviors of autonomous robots to changing or unknown a priori environmental conditions. However, as the size of the space over which a sampling-based approach needs to search is increased (perhaps due to considering robots with many degree of freedom) the computational limits necessary for real-time operation are quickly exceeded. To address this issue, this paper presents a novel sampling-based approach to locomotion planning for highly-articulated robots wherein the parameters associated with a class of locomotive behaviors (e.g., inter-leg coordination, stride length, etc.) are inferred in real-time using a sample-efficient algorithm. More specifically, this work presents a data-based approach wherein offline-learned optimal behaviors, represented using central pattern generators (CPGs), are used to train a class of probabilistic graphical models (PGMs). The trained PGMs are then used to inform a sampling distribution of inferred walking gaits for legged hexapod robots. Simulated as well as hardware results are presented to demonstrate the successful application of the online inference algorithm.more » « less
-
Abstract This paper introduces an adaptive robust trajectory tracking controller design to provably realize stable bipedal robotic walking under parametric and unmodeled uncertainties. Deriving such a controller is challenging mainly because of the highly complex bipedal walking dynamics that are hybrid and involve nonlinear, uncontrolled state-triggered jumps. The main contribution of the study is the synthesis of a continuous-phase adaptive robust tracking control law for hybrid models of bipedal robotic walking by incorporating the construction of multiple Lyapunov functions into the control Lyapunov function. The evolution of the Lyapunov function across the state-triggered jumps is explicitly analyzed to construct sufficient conditions that guide the proposed control design for provably guaranteeing the stability and tracking the performance of the hybrid system in the presence of uncertainties. Simulation results on fully actuated bipedal robotic walking validate the effectiveness of the proposed approach in walking stabilization under uncertainties.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
