Artificial Cerebellum on FPGA: Realistic Real-time Cerebellar Spiking Neural Network Model Capable of Real-world Adaptive Motor Control

Overview

Journal Front Neurosci

Date 2024 May 10

PMID 38726031

Authors

Yusuke Shinji

Hirotsugu Okuno

Yutaka Hirata

Affiliations

Soon will be listed here.

Abstract

The cerebellum plays a central role in motor control and learning. Its neuronal network architecture, firing characteristics of component neurons, and learning rules at their synapses have been well understood in terms of anatomy and physiology. A realistic artificial cerebellum with mimetic network architecture and synaptic plasticity mechanisms may allow us to analyze cerebellar information processing in the real world by applying it to adaptive control of actual machines. Several artificial cerebellums have previously been constructed, but they require high-performance hardware to run in real-time for real-world machine control. Presently, we implemented an artificial cerebellum with the size of 10 spiking neuron models on a field-programmable gate array (FPGA) which is compact, lightweight, portable, and low-power-consumption. In the implementation three novel techniques are employed: (1) 16-bit fixed-point operation and randomized rounding, (2) fully connected spike information transmission, and (3) alternative memory that uses pseudo-random number generators. We demonstrate that the FPGA artificial cerebellum runs in real-time, and its component neuron models behave as those in the corresponding artificial cerebellum configured on a personal computer in Python. We applied the FPGA artificial cerebellum to the adaptive control of a machine in the real world and demonstrated that the artificial cerebellum is capable of adaptively reducing control error after sudden load changes. This is the first implementation and demonstration of a spiking artificial cerebellum on an FPGA applicable to real-world adaptive control. The FPGA artificial cerebellum may provide neuroscientific insights into cerebellar information processing in adaptive motor control and may be applied to various neuro-devices to augment and extend human motor control capabilities.

References

Hirano T . Depression and potentiation of the synaptic transmission between a granule cell and a Purkinje cell in rat cerebellar culture. Neurosci Lett. 1990; 119(2):141-4. DOI: 10.1016/0304-3940(90)90818-t. View

Kuriyama R, Casellato C, DAngelo E, Yamazaki T . Real-Time Simulation of a Cerebellar Scaffold Model on Graphics Processing Units. Front Cell Neurosci. 2021; 15:623552. PMC: 8058369. DOI: 10.3389/fncel.2021.623552. View

Noda H . Mossy fibres sending retinal-slip, eye, and head velocity signals to the flocculus of the monkey. J Physiol. 1986; 379:39-60. PMC: 1182884. DOI: 10.1113/jphysiol.1986.sp016240. View

Fu Y, Djupsund K, Gao H, Hayden B, Shen K, Dan Y . Temporal specificity in the cortical plasticity of visual space representation. Science. 2002; 296(5575):1999-2003. DOI: 10.1126/science.1070521. View

Yang S, Wang J, Zhang N, Deng B, Pang Y, Azghadi M . CerebelluMorphic: Large-Scale Neuromorphic Model and Architecture for Supervised Motor Learning. IEEE Trans Neural Netw Learn Syst. 2021; 33(9):4398-4412. DOI: 10.1109/TNNLS.2021.3057070. View

Miles F, Kawano K, Optican L . Short-latency ocular following responses of monkey. I. Dependence on temporospatial properties of visual input. J Neurophysiol. 1986; 56(5):1321-54. DOI: 10.1152/jn.1986.56.5.1321. View

Allen C, Celikel T, Feldman D . Long-term depression induced by sensory deprivation during cortical map plasticity in vivo. Nat Neurosci. 2003; 6(3):291-9. DOI: 10.1038/nn1012. View

Ekerot C, Jorntell H . Parallel fiber receptive fields: a key to understanding cerebellar operation and learning. Cerebellum. 2003; 2(2):101-9. DOI: 10.1080/14734220309411. View

Mosbacher Y, Khoyratee F, Goldin M, Kanner S, Malakai Y, Silva M . Toward neuroprosthetic real-time communication from in silico to biological neuronal network via patterned optogenetic stimulation. Sci Rep. 2020; 10(1):7512. PMC: 7200763. DOI: 10.1038/s41598-020-63934-4. View

10.

Pinzon-Morales R, Hirata Y . A realistic bi-hemispheric model of the cerebellum uncovers the purpose of the abundant granule cells during motor control. Front Neural Circuits. 2015; 9:18. PMC: 4416449. DOI: 10.3389/fncir.2015.00018. View

11.

Gao Z, van Beugen B, De Zeeuw C . Distributed synergistic plasticity and cerebellar learning. Nat Rev Neurosci. 2012; 13(9):619-35. DOI: 10.1038/nrn3312. View

12.

Seese T, Harasaki H, Saidel G, Davies C . Characterization of tissue morphology, angiogenesis, and temperature in the adaptive response of muscle tissue to chronic heating. Lab Invest. 1999; 78(12):1553-62. View

13.

Luo J, Coapes G, Mak T, Yamazaki T, Tin C, Degenaar P . Real-Time Simulation of Passage-of-Time Encoding in Cerebellum Using a Scalable FPGA-Based System. IEEE Trans Biomed Circuits Syst. 2015; 10(3):742-53. DOI: 10.1109/TBCAS.2015.2460232. View

14.

Monzee J, Drew T, Smith A . Effects of muscimol inactivation of the cerebellar nuclei on precision grip. J Neurophysiol. 2003; 91(3):1240-9. DOI: 10.1152/jn.01124.2002. View

15.

Casali S, Marenzi E, Medini C, Casellato C, DAngelo E . Reconstruction and Simulation of a Scaffold Model of the Cerebellar Network. Front Neuroinform. 2019; 13:37. PMC: 6530631. DOI: 10.3389/fninf.2019.00037. View

16.

Solinas S, Nieus T, DAngelo E . A realistic large-scale model of the cerebellum granular layer predicts circuit spatio-temporal filtering properties. Front Cell Neurosci. 2010; 4:12. PMC: 2876868. DOI: 10.3389/fncel.2010.00012. View

17.

Bi G, Poo M . Synaptic modifications in cultured hippocampal neurons: dependence on spike timing, synaptic strength, and postsynaptic cell type. J Neurosci. 1998; 18(24):10464-72. PMC: 6793365. View

18.

Inagaki K, Hirata Y . Computational Theory Underlying Acute Vestibulo-ocular Reflex Motor Learning with Cerebellar Long-Term Depression and Long-Term Potentiation. Cerebellum. 2017; 16(4):827-839. DOI: 10.1007/s12311-017-0857-6. View

19.

Ito M, Udo M, Mano N, Kawai N . Synaptic action of the fastigiobulbar impulses upon neurones in the medullary reticular formation and vestibular nuclei. Exp Brain Res. 1970; 11(1):29-47. DOI: 10.1007/BF00234200. View

20.

Pinzon-Morales R, Hirata Y . A bi-hemispheric neuronal network model of the cerebellum with spontaneous climbing fiber firing produces asymmetrical motor learning during robot control. Front Neural Circuits. 2014; 8:131. PMC: 4221029. DOI: 10.3389/fncir.2014.00131. View