Vision-Based Control for Robots by a Fully Spiking Neural System Relying on Cerebellar Predictive Learning

Zahra, Omar; Navarro-Alarcon, David; Tolu, Silvia

doi:10.48550/arxiv.2011.01641

Cited by 1 publication

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The training and testing is done one by one for each direction. This outperforms the network developed in [16], as shown in Fig. 9b with a reduction of 55% in the maximum deviation and 120% in execution time, after training for 8 repetitions.…”

Section: A Radial Reachingmentioning

confidence: 81%

“…The aim of this work is to develop a controller based on a cerebellar-like model, developed on the cellular-level, to guide the motion of robots with real-time sensory information. The cerebellar model developed in this study is more detailed from the biological perspective to the previously developed model [16], [17] to demonstrate the effect of these additional features and its effect on the performance. The controller first builds a sensorimotor differential map through motor babbling and the cerebellum acts as a Smith predictor [18] to correct discrepancies in sensory readings to enhance accuracy and precision of robot movements.…”

Section: Introductionmentioning

confidence: 99%

“…Such combination aims to achieve the sensorimotor adaptation, but it does not model the spiking activity in the cerebellum. In [16], a cellular level cerebellar model is developed in which the teaching signal is provided based on the error in task-space. In all the previous studies, no clear method was identified to tune the network parameters, and the developed networks were used to manipulate only the end-effector in some predefined trajectory or to reach a certain target.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Neurorobotic Embodiment for Exploring the Dynamical Interactions of a Spiking Cerebellar Model and a Robot Arm During Vision-Based Manipulation Tasks

Zahra

Navarro-Alarcon

Tolu

2021

Int. J. Neur. Syst.

Self Cite

View full text Add to dashboard Cite

While the original goal for developing robots is replacing humans in dangerous and tedious tasks, the final target shall be completely mimicking the human cognitive and motor behavior. Hence, building detailed computational models for the human brain is one of the reasonable ways to attain this. The cerebellum is one of the key players in our neural system to guarantee dexterous manipulation and coordinated movements as concluded from lesions in that region. Studies suggest that it acts as a forward model providing anticipatory corrections for the sensory signals based on observed discrepancies from the reference values. While most studies consider providing the teaching signal as error in joint-space, few studies consider the error in task-space and even fewer consider the spiking nature of the cerebellum on the cellular-level. In this study, a detailed cellular-level forward cerebellar model is developed, including modeling of Golgi and Basket cells which are usually neglected in previous studies. To preserve the biological features of the cerebellum in the developed model, a hyperparameter optimization method tunes the network accordingly. The efficiency and biological plausibility of the proposed cerebellar-based controller is then demonstrated under different robotic manipulation tasks reproducing motor behavior observed in human reaching experiments.

show abstract

Section: A Radial Reachingmentioning

confidence: 81%