Skip to main content

Silvia Tolu

Technical University of Denmark (DTU), Department of Electrical Engineering, Adjunct

Followers

46

Following

28

Public Views

I am employed as Assistant Professor at the Automation and Control group - DTU Electrical Engineering. My research at DTU is in the field of Neuro-robotics Engineering, in particular it is focused on an understanding of a modular approach for developing brain structures and to connect them for controlling different robotic platforms, e.g. modular robots, musculoskeletal robots, humanoids robots, and soft-robots. Most of my research has been performed in the context of European Projects (Sensopac, HBP, BioModular). Current research interests are on the following areas: Neuro-robotics Neuromorphic Engineering Bio-inspired control systems Artificial Intelligence Applied Computational Neuroscience

less

University of California, Santa Barbara

Harvard University

University of California, Merced

Kansas State University

Ute Römer-Barron

Georgia State University

Armando Marques-Guedes

UNL - New University of Lisbon

The Hebrew University of Jerusalem

Graduate Center of the City University of New York

Indiana University

Viacheslav Kuleshov

Stockholm University

InterestsView All (11)

Uploads

Papers by Silvia Tolu

A Learning-based Control Framework for Fast and Accurate Manipulation of a Flexible Object

Journal of bionic engineering/Journal of Bionics Engineering, May 21, 2024

This paper presents a learning-based control framework for fast (< 1.5 s) and accurate manipulati... more This paper presents a learning-based control framework for fast (< 1.5 s) and accurate manipulation of a flexible object, i.e., whip targeting. The framework consists of a motion planner learned or optimized by an algorithm, Online Impedance Adaptation Control (OIAC), a sim2real mechanism, and a visual feedback component. The experimental results show that a soft actor-critic algorithm outperforms three Deep Reinforcement Learning (DRL), a nonlinear optimization, and a genetic algorithm in learning generalization of motion planning. It can greatly reduce average learning trials (to < 20% of others) and maximize average rewards (to > 3 times of others). Besides, motion tracking errors are greatly reduced to 13.29% and 22.36% of constant impedance control by the OIAC of the proposed framework. In addition, the trajectory similarity between simulated and physical whips is 89.09% . The presented framework provides a new method integrating data-driven and physics-based algorithms for controlling fast and accurate arm manipulation of a flexible object.

Editorial: Women in neurorobotics

Frontiers in Neurorobotics, Nov 27, 2023

Sliding-mode control of a soft robot based on data-driven sparse identification

Control Engineering Practice, Feb 29, 2024

De cine: una secuencia didáctica para niveles B1-B2

Azulejo para el aula de español, 2013

Motor Learning for Manoeuvring Control of a Remotely Operated Vehicle

IFAC-PapersOnLine, 2022

Proceedings of the Neuro-inspired Computational Elements Workshop

Human beings and other vertebrates show remarkable performance and efficiency in locomotion, but ... more Human beings and other vertebrates show remarkable performance and efficiency in locomotion, but the functioning of their biological control systems for locomotion is still only partially understood. The basic patterns and timing for locomotion are provided by a central pattern generator (CPG) in the spinal cord. The cerebellum is known to play an important role in adaptive locomotion. Recent studies have given insights into the error signals responsible for driving the cerebellar adaptation in locomotion. However, the question of how the cerebellar output influences the gait remains unanswered. We hypothesize that the cerebellar correction is applied to the pattern formation part of the CPG. Here, a bio-inspired control system for adaptive locomotion of the musculoskeletal system of the mouse is presented, where a cerebellar-like module adapts the step time by using the double support interlimb asymmetry as a temporal teaching signal. The control system is tested on a simulated mouse in a split-belt treadmill setup similar to those used in experiments with real mice. The results show adaptive locomotion behavior in the interlimb parameters similar to that seen in humans and mice. The control system adaptively decreases the double support asymmetry that occurs due to environmental perturbations in the split-belt protocol.

Perspective on investigation of neurodegenerative diseases with neurorobotics approaches

Neuromorphic Computing and Engineering

Neurorobotics has emerged from the alliance between neuroscience and robotics. It pursues the inv... more Neurorobotics has emerged from the alliance between neuroscience and robotics. It pursues the investigation of reproducing living organism-like behaviors in robots by means of the embodiment of computational models of the central nervous system. This perspective article discusses the current trend of implementing tools for the pressing challenge of early-diagnosis of neurodegenerative diseases and how neurorobotics approaches can help. Recently, advances in this field have allowed the testing of some neuroscientific hypotheses related to brain diseases, but the lack of biological plausibility of developed brain models and musculoskeletal systems has limited the understanding of the underlying brain mechanisms that lead to deficits in motor and cognitive tasks. Key aspects and methods to enhance the reproducibility of natural behaviors observed in healthy and impaired brains are proposed in this perspective. In the long term, the goal is to move beyond finding therapies and look into...

A Bio-inspired Mechanism for Learning from Mirrored Demonstrations

2 Different learning modes and mechanisms allow faster and better acquisition of skills as widely... more 2 Different learning modes and mechanisms allow faster and better acquisition of skills as widely 3 studied in humans and many animals. Specific neurons, called mirror neurons, are activated in 4 the same way whether an action is performed or simply observed. This suggests that observing 5 others performing movements allows to reinforce our motor abilities. This implies the presence 6 of a biological mechanism that allows creating models of others’ movements and linking them 7 to the self-model for achieving mirroring. Inspired by such ability, we propose to build a map 8 of movements executed by a teaching agent and mirroring the agent’s state to the robot’s 9 configuration space. Hence, in this study, a neural network is proposed to integrate a motor cortex10 like differential map transforming motor plans from task-space to joint-space motor commands 11 and a static map correlating joint-spaces of the robot and a teaching agent. The differential map 12 is developed based on spikin...

Editorial: Biomimetic control architectures for robots

Frontiers in Neurorobotics

A Bio-Inspired Mechanism for Learning Robot Motion From Mirrored Human Demonstrations

Frontiers in Neurorobotics, 2022

Different learning modes and mechanisms allow faster and better acquisition of skills as widely s... more Different learning modes and mechanisms allow faster and better acquisition of skills as widely studied in humans and many animals. Specific neurons, called mirror neurons, are activated in the same way whether an action is performed or simply observed. This suggests that observing others performing movements allows to reinforce our motor abilities. This implies the presence of a biological mechanism that allows creating models of others' movements and linking them to the self-model for achieving mirroring. Inspired by such ability, we propose to build a map of movements executed by a teaching agent and mirror the agent's state to the robot's configuration space. Hence, in this study, a neural network is proposed to integrate a motor cortex-like differential map transforming motor plans from task-space to joint-space motor commands and a static map correlating joint-spaces of the robot and a teaching agent. The differential map is developed based on spiking neural networ...

Closed-loop Dynamic Control of a Soft Manipulator using Deep Reinforcement Learning

IEEE Robotics and Automation Letters, 2022

Adaptive Cerebellar Spiking Model Embedded in the Control Loop: Context Switching and Robustness Against Noise

International Journal of Neural Systems, 2011

This work evaluates the capability of a spiking cerebellar model embedded in different loop archi... more This work evaluates the capability of a spiking cerebellar model embedded in different loop architectures (recurrent, forward, and forward&amp;recurrent) to control a robotic arm (three degrees of freedom) using a biologically-inspired approach. The implemented spiking network relies on synaptic plasticity (long-term potentiation and long-term depression) to adapt and cope with perturbations in the manipulation scenario: changes in dynamics and kinematics of the simulated robot. Furthermore, the effect of several degrees of noise in the cerebellar input pathway (mossy fibers) was assessed depending on the employed control architecture. The implemented cerebellar model managed to adapt in the three control architectures to different dynamics and kinematics providing corrective actions for more accurate movements. According to the obtained results, coupling both control architectures (forward&amp;recurrent) provides benefits of the two of them and leads to a higher robustness against noise.

Adaptive and Predictive Control of a Simulated Robot Arm

International Journal of Neural Systems, 2013

In this work, a basic cerebellar neural layer and a machine learning engine are embedded in a rec... more In this work, a basic cerebellar neural layer and a machine learning engine are embedded in a recurrent loop which avoids dealing with the motor error or distal error problem. The presented approach learns the motor control based on available sensor error estimates (position, velocity, and acceleration) without explicitly knowing the motor errors. The paper focuses on how to decompose the input into different components in order to facilitate the learning process using an automatic incremental learning model (locally weighted projection regression (LWPR) algorithm). LWPR incrementally learns the forward model of the robot arm and provides the cerebellar module with optimal pre-processed signals. We present a recurrent adaptive control architecture in which an adaptive feedback (AF) controller guarantees a precise, compliant, and stable control during the manipulation of objects. Therefore, this approach efficiently integrates a bio-inspired module (cerebellar circuitry) with a machi...

Dynamics Model Abstraction Scheme Using Radial Basis Functions

Journal of Control Science and Engineering, 2012

This paper presents a control model for object manipulation. Properties of objects and environmen... more This paper presents a control model for object manipulation. Properties of objects and environmental conditions influence the motor control and learning. System dynamics depend on an unobserved external context, for example, work load of a robot manipulator. The dynamics of a robot arm change as it manipulates objects with different physical properties, for example, the mass, shape, or mass distribution. We address active sensing strategies to acquire object dynamical models with a radial basis function neural network (RBF). Experiments are done using a real robot’s arm, and trajectory data are gathered during various trials manipulating different objects. Biped robots do not have high force joint servos and the control system hardly compensates all the inertia variation of the adjacent joints and disturbance torque on dynamic gait control. In order to achieve smoother control and lead to more reliable sensorimotor complexes, we evaluate and compare a sparse velocity-driven versus a...

Bio-inspired adaptive feedback error learning architecture for motor control

Biological Cybernetics, 2012

This study proposes an adaptive control architecture based on an accurate regression method calle... more This study proposes an adaptive control architecture based on an accurate regression method called Locally Weighted Projection Regression (LWPR) and on a bio-inspired module, such as a cerebellar-like engine. This hybrid architecture takes full advantage of the machine learning module (LWPR kernel) to abstract an optimized representation of the sensorimotor space while the cerebellar component integrates this to generate corrective terms in the framework of a control task. Furthermore, we illustrate how the use of a simple adaptive error feedback term allows to use the proposed architecture even in the absence of an accurate analytic reference model. The presented approach achieves an accurate control with low gain corrective terms (for compliant control schemes). We evaluate the contribution of the different components of the proposed scheme comparing the obtained performance with alternative approaches. Then, we show that the presented architecture can be used for accurate manipulation of different objects when their physical properties are not directly known by the controller. We evaluate how the scheme scales for simulated plants of high Degrees of Freedom (7-DOFs).

Distributed and Modular Bio-Inspired Architecture for Adaptive Motor Learning and Control

Open-loop Model-free Dynamic Control of a Soft Manipulator for Tracking Tasks

2021 20th International Conference on Advanced Robotics (ICAR)

Adaptive Variable Impedance Control for a Modular Soft Robot Manipulator in Configuration Space

ArXiv, 2021

Compliance is a strong requirement for human-robot interactions. Soft-robots provide an opportuni... more Compliance is a strong requirement for human-robot interactions. Soft-robots provide an opportunity to cover the lack of compliance in conventional actuation mechanisms, however, the control of them is very challenging given their intrinsic complex motions. Therefore, soft-robots require new approaches to e.g., modeling, control, dynamics, and planning. One of the control strategies that ensures compliance is the impedance control. During the task execution in the presence of coupling force and position constraints, a dynamic behavior increases the flexibility of the impedance control. This imposes some additional constraints on the stability of the control system. To tackle them, we propose a variable impedance control in configuration space for a modular soft robot manipulator (MSRM) in the presence of model uncertainties and external forces. The external loads are estimated in configuration space using a momentum-based approach in order to reduce the calculation complexity, and t...

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

ArXiv, 2020

The cerebellum plays a distinctive role within our motor control system to achieve fine and coord... more The cerebellum plays a distinctive role within our motor control system to achieve fine and coordinated motions. While cerebellar lesions do not lead to a complete loss of motor functions, both action and perception are severally impacted. Hence, it is assumed that the cerebellum uses an internal forward model to provide anticipatory signals by learning from the error in sensory states. In some studies, it was demonstrated that the learning process relies on the joint-space error. However, this may not exist. This work proposes a novel fully spiking neural system that relies on a forward predictive learning by means of a cellular cerebellar model. The forward model is learnt thanks to the sensory feedback in task-space and it acts as a Smith predictor. The latter predicts sensory corrections in input to a differential mapping spiking neural network during a visual servoing task of a robot arm manipulator. In this paper, we promote the developed control system to achieve more accurat...

2022 Roadmap on Neuromorphic Computing and Engineering

Modern computation based on the von Neumann architecture is today a mature cutting-edge science. ... more Modern computation based on the von Neumann architecture is today a mature cutting-edge science. In the Von Neumann architecture, processing and memory units are implemented as separate blocks interchanging data intensively and continuously. This data transfer is responsible for a large part of the power consumption. The next generation computer technology is expected to solve problems at the exascale with 1018 calculations each second. Even though these future computers will be incredibly powerful, if they are based on von Neumann type architectures, they will consume between 20 and 30 megawatts of power and will not have intrinsic physically built-in capabilities to learn or deal with complex data as our brain does. These needs can be addressed by neuromorphic computing systems which are inspired by the biological concepts of the human brain. This new generation of computers has the potential to be used for the storage and processing of large amounts of digital information with mu...

A Learning-based Control Framework for Fast and Accurate Manipulation of a Flexible Object

Journal of bionic engineering/Journal of Bionics Engineering, May 21, 2024

This paper presents a learning-based control framework for fast (< 1.5 s) and accurate manipulati... more This paper presents a learning-based control framework for fast (< 1.5 s) and accurate manipulation of a flexible object, i.e., whip targeting. The framework consists of a motion planner learned or optimized by an algorithm, Online Impedance Adaptation Control (OIAC), a sim2real mechanism, and a visual feedback component. The experimental results show that a soft actor-critic algorithm outperforms three Deep Reinforcement Learning (DRL), a nonlinear optimization, and a genetic algorithm in learning generalization of motion planning. It can greatly reduce average learning trials (to < 20% of others) and maximize average rewards (to > 3 times of others). Besides, motion tracking errors are greatly reduced to 13.29% and 22.36% of constant impedance control by the OIAC of the proposed framework. In addition, the trajectory similarity between simulated and physical whips is 89.09% . The presented framework provides a new method integrating data-driven and physics-based algorithms for controlling fast and accurate arm manipulation of a flexible object.

Editorial: Women in neurorobotics

Frontiers in Neurorobotics, Nov 27, 2023

Sliding-mode control of a soft robot based on data-driven sparse identification

Control Engineering Practice, Feb 29, 2024

De cine: una secuencia didáctica para niveles B1-B2

Azulejo para el aula de español, 2013

Motor Learning for Manoeuvring Control of a Remotely Operated Vehicle

IFAC-PapersOnLine, 2022

Proceedings of the Neuro-inspired Computational Elements Workshop

Human beings and other vertebrates show remarkable performance and efficiency in locomotion, but ... more Human beings and other vertebrates show remarkable performance and efficiency in locomotion, but the functioning of their biological control systems for locomotion is still only partially understood. The basic patterns and timing for locomotion are provided by a central pattern generator (CPG) in the spinal cord. The cerebellum is known to play an important role in adaptive locomotion. Recent studies have given insights into the error signals responsible for driving the cerebellar adaptation in locomotion. However, the question of how the cerebellar output influences the gait remains unanswered. We hypothesize that the cerebellar correction is applied to the pattern formation part of the CPG. Here, a bio-inspired control system for adaptive locomotion of the musculoskeletal system of the mouse is presented, where a cerebellar-like module adapts the step time by using the double support interlimb asymmetry as a temporal teaching signal. The control system is tested on a simulated mouse in a split-belt treadmill setup similar to those used in experiments with real mice. The results show adaptive locomotion behavior in the interlimb parameters similar to that seen in humans and mice. The control system adaptively decreases the double support asymmetry that occurs due to environmental perturbations in the split-belt protocol.

Perspective on investigation of neurodegenerative diseases with neurorobotics approaches

Neuromorphic Computing and Engineering

Neurorobotics has emerged from the alliance between neuroscience and robotics. It pursues the inv... more Neurorobotics has emerged from the alliance between neuroscience and robotics. It pursues the investigation of reproducing living organism-like behaviors in robots by means of the embodiment of computational models of the central nervous system. This perspective article discusses the current trend of implementing tools for the pressing challenge of early-diagnosis of neurodegenerative diseases and how neurorobotics approaches can help. Recently, advances in this field have allowed the testing of some neuroscientific hypotheses related to brain diseases, but the lack of biological plausibility of developed brain models and musculoskeletal systems has limited the understanding of the underlying brain mechanisms that lead to deficits in motor and cognitive tasks. Key aspects and methods to enhance the reproducibility of natural behaviors observed in healthy and impaired brains are proposed in this perspective. In the long term, the goal is to move beyond finding therapies and look into...

A Bio-inspired Mechanism for Learning from Mirrored Demonstrations

2 Different learning modes and mechanisms allow faster and better acquisition of skills as widely... more 2 Different learning modes and mechanisms allow faster and better acquisition of skills as widely 3 studied in humans and many animals. Specific neurons, called mirror neurons, are activated in 4 the same way whether an action is performed or simply observed. This suggests that observing 5 others performing movements allows to reinforce our motor abilities. This implies the presence 6 of a biological mechanism that allows creating models of others’ movements and linking them 7 to the self-model for achieving mirroring. Inspired by such ability, we propose to build a map 8 of movements executed by a teaching agent and mirroring the agent’s state to the robot’s 9 configuration space. Hence, in this study, a neural network is proposed to integrate a motor cortex10 like differential map transforming motor plans from task-space to joint-space motor commands 11 and a static map correlating joint-spaces of the robot and a teaching agent. The differential map 12 is developed based on spikin...

Editorial: Biomimetic control architectures for robots

Frontiers in Neurorobotics

A Bio-Inspired Mechanism for Learning Robot Motion From Mirrored Human Demonstrations

Frontiers in Neurorobotics, 2022

Different learning modes and mechanisms allow faster and better acquisition of skills as widely s... more Different learning modes and mechanisms allow faster and better acquisition of skills as widely studied in humans and many animals. Specific neurons, called mirror neurons, are activated in the same way whether an action is performed or simply observed. This suggests that observing others performing movements allows to reinforce our motor abilities. This implies the presence of a biological mechanism that allows creating models of others' movements and linking them to the self-model for achieving mirroring. Inspired by such ability, we propose to build a map of movements executed by a teaching agent and mirror the agent's state to the robot's configuration space. Hence, in this study, a neural network is proposed to integrate a motor cortex-like differential map transforming motor plans from task-space to joint-space motor commands and a static map correlating joint-spaces of the robot and a teaching agent. The differential map is developed based on spiking neural networ...

Closed-loop Dynamic Control of a Soft Manipulator using Deep Reinforcement Learning

IEEE Robotics and Automation Letters, 2022

Adaptive Cerebellar Spiking Model Embedded in the Control Loop: Context Switching and Robustness Against Noise

International Journal of Neural Systems, 2011

This work evaluates the capability of a spiking cerebellar model embedded in different loop archi... more This work evaluates the capability of a spiking cerebellar model embedded in different loop architectures (recurrent, forward, and forward&amp;recurrent) to control a robotic arm (three degrees of freedom) using a biologically-inspired approach. The implemented spiking network relies on synaptic plasticity (long-term potentiation and long-term depression) to adapt and cope with perturbations in the manipulation scenario: changes in dynamics and kinematics of the simulated robot. Furthermore, the effect of several degrees of noise in the cerebellar input pathway (mossy fibers) was assessed depending on the employed control architecture. The implemented cerebellar model managed to adapt in the three control architectures to different dynamics and kinematics providing corrective actions for more accurate movements. According to the obtained results, coupling both control architectures (forward&amp;recurrent) provides benefits of the two of them and leads to a higher robustness against noise.

Adaptive and Predictive Control of a Simulated Robot Arm

International Journal of Neural Systems, 2013

In this work, a basic cerebellar neural layer and a machine learning engine are embedded in a rec... more In this work, a basic cerebellar neural layer and a machine learning engine are embedded in a recurrent loop which avoids dealing with the motor error or distal error problem. The presented approach learns the motor control based on available sensor error estimates (position, velocity, and acceleration) without explicitly knowing the motor errors. The paper focuses on how to decompose the input into different components in order to facilitate the learning process using an automatic incremental learning model (locally weighted projection regression (LWPR) algorithm). LWPR incrementally learns the forward model of the robot arm and provides the cerebellar module with optimal pre-processed signals. We present a recurrent adaptive control architecture in which an adaptive feedback (AF) controller guarantees a precise, compliant, and stable control during the manipulation of objects. Therefore, this approach efficiently integrates a bio-inspired module (cerebellar circuitry) with a machi...

Dynamics Model Abstraction Scheme Using Radial Basis Functions

Journal of Control Science and Engineering, 2012

This paper presents a control model for object manipulation. Properties of objects and environmen... more This paper presents a control model for object manipulation. Properties of objects and environmental conditions influence the motor control and learning. System dynamics depend on an unobserved external context, for example, work load of a robot manipulator. The dynamics of a robot arm change as it manipulates objects with different physical properties, for example, the mass, shape, or mass distribution. We address active sensing strategies to acquire object dynamical models with a radial basis function neural network (RBF). Experiments are done using a real robot’s arm, and trajectory data are gathered during various trials manipulating different objects. Biped robots do not have high force joint servos and the control system hardly compensates all the inertia variation of the adjacent joints and disturbance torque on dynamic gait control. In order to achieve smoother control and lead to more reliable sensorimotor complexes, we evaluate and compare a sparse velocity-driven versus a...

Bio-inspired adaptive feedback error learning architecture for motor control

Biological Cybernetics, 2012

This study proposes an adaptive control architecture based on an accurate regression method calle... more This study proposes an adaptive control architecture based on an accurate regression method called Locally Weighted Projection Regression (LWPR) and on a bio-inspired module, such as a cerebellar-like engine. This hybrid architecture takes full advantage of the machine learning module (LWPR kernel) to abstract an optimized representation of the sensorimotor space while the cerebellar component integrates this to generate corrective terms in the framework of a control task. Furthermore, we illustrate how the use of a simple adaptive error feedback term allows to use the proposed architecture even in the absence of an accurate analytic reference model. The presented approach achieves an accurate control with low gain corrective terms (for compliant control schemes). We evaluate the contribution of the different components of the proposed scheme comparing the obtained performance with alternative approaches. Then, we show that the presented architecture can be used for accurate manipulation of different objects when their physical properties are not directly known by the controller. We evaluate how the scheme scales for simulated plants of high Degrees of Freedom (7-DOFs).

Distributed and Modular Bio-Inspired Architecture for Adaptive Motor Learning and Control

Open-loop Model-free Dynamic Control of a Soft Manipulator for Tracking Tasks

2021 20th International Conference on Advanced Robotics (ICAR)

Adaptive Variable Impedance Control for a Modular Soft Robot Manipulator in Configuration Space

ArXiv, 2021

Compliance is a strong requirement for human-robot interactions. Soft-robots provide an opportuni... more Compliance is a strong requirement for human-robot interactions. Soft-robots provide an opportunity to cover the lack of compliance in conventional actuation mechanisms, however, the control of them is very challenging given their intrinsic complex motions. Therefore, soft-robots require new approaches to e.g., modeling, control, dynamics, and planning. One of the control strategies that ensures compliance is the impedance control. During the task execution in the presence of coupling force and position constraints, a dynamic behavior increases the flexibility of the impedance control. This imposes some additional constraints on the stability of the control system. To tackle them, we propose a variable impedance control in configuration space for a modular soft robot manipulator (MSRM) in the presence of model uncertainties and external forces. The external loads are estimated in configuration space using a momentum-based approach in order to reduce the calculation complexity, and t...

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

ArXiv, 2020

The cerebellum plays a distinctive role within our motor control system to achieve fine and coord... more The cerebellum plays a distinctive role within our motor control system to achieve fine and coordinated motions. While cerebellar lesions do not lead to a complete loss of motor functions, both action and perception are severally impacted. Hence, it is assumed that the cerebellum uses an internal forward model to provide anticipatory signals by learning from the error in sensory states. In some studies, it was demonstrated that the learning process relies on the joint-space error. However, this may not exist. This work proposes a novel fully spiking neural system that relies on a forward predictive learning by means of a cellular cerebellar model. The forward model is learnt thanks to the sensory feedback in task-space and it acts as a Smith predictor. The latter predicts sensory corrections in input to a differential mapping spiking neural network during a visual servoing task of a robot arm manipulator. In this paper, we promote the developed control system to achieve more accurat...

2022 Roadmap on Neuromorphic Computing and Engineering

Modern computation based on the von Neumann architecture is today a mature cutting-edge science. ... more Modern computation based on the von Neumann architecture is today a mature cutting-edge science. In the Von Neumann architecture, processing and memory units are implemented as separate blocks interchanging data intensively and continuously. This data transfer is responsible for a large part of the power consumption. The next generation computer technology is expected to solve problems at the exascale with 1018 calculations each second. Even though these future computers will be incredibly powerful, if they are based on von Neumann type architectures, they will consume between 20 and 30 megawatts of power and will not have intrinsic physically built-in capabilities to learn or deal with complex data as our brain does. These needs can be addressed by neuromorphic computing systems which are inspired by the biological concepts of the human brain. This new generation of computers has the potential to be used for the storage and processing of large amounts of digital information with mu...