Neuromorphic Robotic Platform with Visual Input, Processor and Actuator, Based on Spiking Neural Networks

This paper describes the design and modus of operation of a neuromorphic robotic platform based on SpiNNaker, and its implementation on the goalkeeper task. The robotic system utilises an address event representation (AER) type of camera (dynamic vision sensor (DVS)) to capture features of a moving ball, and a servo motor to position the goalkeeper to intercept the incoming ball. At the backbone of the system is a microcontroller (Arduino Due) which facilitates communication and control between different robot parts. A spiking neuronal network (SNN), which is running on SpiNNaker, predicts the location of arrival of the moving ball and decides where to place the goalkeeper. In our setup, the maximum data transmission speed of the closed-loop system is approximately 3000 packets per second for both uplink and downlink, and the robot can intercept balls whose speed is up to 1 m/s starting from the distance of about 0.8 m. The interception accuracy is up to 85%, the response latency is 6.5 ms and the maximum power consumption is 7.15 W. This is better than previous implementations based on PC. Here, a simplified version of an SNN has been developed for the ‘interception of a moving object’ task, for the purpose of demonstrating the platform, however a generalised SNN for this problem is a nontrivial problem. A demo video of the robot goalie is available on YouTube.

Download Full-text

System integration of neuromorphic visual (DVS), processing (SpiNNaker) and servomotor modules into an autonomous robot controlled by a Spiking Neural Network

10.36227/techrxiv.11854254.v1 ◽

2020 ◽

Author(s):

Ran Cheng ◽

Konstantin Nikolic

Keyword(s):

Neural Network ◽

Data Transmission ◽

System Integration ◽

Closed Loop ◽

Robotic System ◽

Spiking Neural Network ◽

Servo Motor ◽

Event Representation ◽

Closed Loop System ◽

Transmission Speed

<p>This paper describes the design and modus of operation of a neuromorphic robotic system based on SpiNNaker platform, and its implementation on the goalkeeper task. The robotic system utilizes an Address Event Representation (AER) type of camera (DVS) to capture features of a moving ball, and a servo motor to position the goalkeeper to intercept the incoming ball. At the backbone of the system is a microprocessor (Arduino Due) which facilitates the communication between different robot parts. A spiking neural network, which is running on SpiNNaker, predicts the location of arrival of the moving ball and decides where to place the goalkeeper. In our setup, the maximum data transmission speed of the closed-loop system is approximately 3000 packets per second for both uplink and downlink, and the robot can intercept balls whose speed is up to 1 m/s starting from the distance of about 0.8 m. The interception accuracy is 85%, the response latency is 6.5 ms and the maximum power consumption is 7.15 W. A demo video of the robot goalie is available on YouTube: <a href="https://www.youtube.com/watch?v=135fH21QXtg">https://www.youtube.com/watch?v=135fH21QXtg</a> .</p>

Download Full-text

System integration of neuromorphic visual (DVS), processing (SpiNNaker) and servomotor modules into an autonomous robot controlled by a Spiking Neural Network

10.36227/techrxiv.11854254 ◽

2020 ◽

Author(s):

Ran Cheng ◽

Konstantin Nikolic

Keyword(s):

Neural Network ◽

Data Transmission ◽

System Integration ◽

Closed Loop ◽

Robotic System ◽

Spiking Neural Network ◽

Servo Motor ◽

Event Representation ◽

Closed Loop System ◽

Transmission Speed

Download Full-text

Design, Construction, and Control for an Underwater Vehicle Type Sepiida

Robotica ◽

10.1017/s0263574720000739 ◽

2020 ◽

pp. 1-18

Author(s):

M. Garcia ◽

P. Castillo ◽

E. Campos ◽

R. Lozano

Keyword(s):

Embedded System ◽

Closed Loop ◽

Underwater Vehicle ◽

Mathematical Representation ◽

Closed Loop System ◽

Vehicle Type ◽

Mathematical Equations ◽

And Control ◽

Design Construction

SUMMARY A novel underwater vehicle configuration with an operating principle as the Sepiida animal is presented and developed in this paper. The mathematical equations describing the movements of the vehicle are obtained using the Newton–Euler approach. An analysis of the dynamic model is done for control purposes. A prototype and its embedded system are developed for validating analytically and experimentally the proposed mathematical representation. A real-time characterization of one mass is done to relate the pitch angle with the radio of displacement of the mass. In addition, first validation of the closed-loop system is done using a linear controller.

Download Full-text

Dynamic Modeling and Control of Packing Pressure in Injection Molding

Journal of Engineering Materials and Technology ◽

10.1115/1.2904280 ◽

1994 ◽

Vol 116 (2) ◽

pp. 244-249 ◽

Cited By ~ 8

Author(s):

J. Hu ◽

J. H. Vogel

Keyword(s):

Injection Molding ◽

Closed Loop ◽

Good Control ◽

Pressure Control ◽

Physical Parameters ◽

Closed Loop System ◽

Modeling And Control ◽

Packing Pressure ◽

And Control ◽

Plastic Injection

A dynamic model of injection molding developed from physical considerations is used to select PID gains for pressure control during the packing phase of thermo-plastic injection molding. The relative importance of various aspects of the model and values for particular physical parameters were identified experimentally. The controller gains were chosen by pole-zero cancellation and root-locus methods, resulting in good control performance. Both open and closed-loop system responses were predicted and verified, with good overall agreement.

Download Full-text

Design of Output Feedback Controller for Switched Fuzzy Systems with Time-Delay

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.635-637.1443 ◽

2014 ◽

Vol 635-637 ◽

pp. 1443-1446

Author(s):

Hong Yang ◽

Huan Huan Lü ◽

Le Zhang

Keyword(s):

Time Delay ◽

Output Feedback ◽

Fuzzy Systems ◽

Closed Loop ◽

Feedback Controller ◽

Linear Matrix ◽

Sufficient Condition ◽

Closed Loop System ◽

Output Feedback Controller ◽

And Control

This paper investigates the problems of stabilization and control for time-delay switched fuzzy systems using output feedback controller. Based on the linear matrix inequality (LMI) technique, multiple Lyapunov method is used to obtain a sufficient condition for the existence of the controller for the output feedback. Then an algorithm is constructed to transform the sufficient condition into a LMI form, thus obtaining a method for designing the controller. The designed controller guarantees the closed-loop system to be asympototically stable. A numerical example is given to show the effectiveness of our method.

Download Full-text

Feedback Control of Linear Distributed Systems

Journal of Basic Engineering ◽

10.1115/1.3609611 ◽

1967 ◽

Vol 89 (2) ◽

pp. 379-383 ◽

Cited By ~ 16

Author(s):

Donald M. Wiberg

Keyword(s):

Boundary Condition ◽

Distributed Systems ◽

Feedback Control ◽

Closed Loop ◽

The State ◽

Loop System ◽

Closed Loop System ◽

Loop Equation ◽

Control Feedback ◽

And Control

The optimum feedback control of controllable linear distributed stationary systems is discussed. A linear closed-loop system is assured by restricting the criterion to be the integral of quadratics in the state and control. Feedback is obtained by expansion of the linear closed-loop equation in terms of uncoupled modes. By incorporating symbolic functions into the formulation, one can treat boundary condition control and point observable systems that are null-delta controllable.

Download Full-text

Multi-Objective Robust Decentralized Control for the Interconnected Fuzzy Singularly Perturbed Models

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.182-183.1200 ◽

2012 ◽

Vol 182-183 ◽

pp. 1200-1205

Author(s):

Ye Nan Hu ◽

Fu Chun Sun

Keyword(s):

Decentralized Control ◽

Closed Loop ◽

Control Method ◽

Dynamic Performance ◽

Singularly Perturbed ◽

Disturbance Attenuation ◽

Closed Loop System ◽

Multi Objective ◽

Robust Decentralized Control ◽

And Control

A multi-objective robust decentralized control method is proposed for the interconnected fuzzy singularly perturbed models. Such decentralized controller can guarantee the whole closed-loop system is asymptotically stable even when the multi-time-scale subsystems are interactional. Besides, the disturbance attenuation performance, dynamic performance and control amplitude can be optimized synthetically. The simulations illustrate the effectiveness of the proposed method.

Download Full-text

Modeling and Control of an Automotive All-Wheel Drive Clutch as a Piecewise Affine System

Journal of Dynamic Systems Measurement and Control ◽

10.1115/1.4025275 ◽

2013 ◽

Vol 136 (1) ◽

Cited By ~ 1

Author(s):

Shiming Duan ◽

Jun Ni ◽

A. Galip Ulsoy

Keyword(s):

Closed Loop ◽

Open Loop ◽

Lyapunov Theory ◽

Loop System ◽

Linear Matrix ◽

Closed Loop System ◽

Modeling And Control ◽

Piecewise Affine ◽

Wheel Drive ◽

And Control

Piecewise affine (PWA) systems belong to a subclass of switched systems and provide good flexibility and traceability for modeling a variety of nonlinear systems. In this paper, application of the PWA system framework to the modeling and control of an automotive all-wheel drive (AWD) clutch system is presented. The open-loop system is first modeled as a PWA system, followed by the design of a piecewise linear (i.e., switched) feedback controller. The stability of the closed-loop system, including model uncertainty and time delays, is examined using linear matrix inequalities based on Lyapunov theory. Finally, the responses of the closed-loop system under step and sine reference signals and temperature disturbance signals are simulated to illustrate the effectiveness of the design.

Download Full-text

Operation of the auditory feedback monitoring loop in children with articulatory defects

South African Journal of Communication Disorders ◽

10.4102/sajcd.v18i1.422 ◽

1971 ◽

Vol 18 (1) ◽

Author(s):

Sandra Ossip

Keyword(s):

Control System ◽

Motor Activity ◽

Auditory Feedback ◽

Closed Loop ◽

Strong Relationship ◽

Preliminary Evaluation ◽

Closed Loop System ◽

Feedback Monitoring ◽

And Control ◽

Normal Speech

This study constitutes a preliminary evaluation of the utilization of auditory feedback for the acquisition of normal speech in normal speaking children and children having functional articulatory errors. The degree to which this is utilized for the organization and control of motor activity was inferred by delaying auditory feedback in time and quantitating the resulting disturbances in the speech behaviour. Evidence was found to support the following hypotheses: I. There is a breakdown of speech under DAF. 2: Children with multiple articulatory disorders exhibit less severe breakdown effects under DAF than their normal peers. 3. There appears to be a strong relationship between increasing age and articulatory ability. 4. There tends to be a relationship between increasing age and the breakdown of speech under DAF. 5. Monitoring of speech is a highly skilled control system which tends to develop with age and experience, and is not operating as strongly in the child with articulation-defects. From the results of the study, it seems that the auditory feedback monitoring loop for speech is not operating as successfully in the child with multiple articulatory errors as it operates in the normal child, and that the development of a closed loop system appears to be retarded in some way.

Download Full-text