Neuromorphic Control Research Articles

A light-weight neuromorphic controlling clock gating based multi-core cryptography platform

While speeding up cryptography tasks can be accomplished by using a multi-core architecture to parallelize computation, one of the major challenges is optimizing power consumption. In principle, depending on the computation workload, individual cores can be turned off to save power during operation. However, too few active cores may lead to computational bottlenecks. In this work, we propose a novel platform named Spike-MCryptCores: a low-power multi-core AES platform with a neuromorphic controller. The proposed Spike-MCryptCores platform is composed of multiple AES cores, each core is equipped with a clock-gating scheme for reducing its power consumption while being idle. To optimize the power consumption of the whole platform, we use a neuromorphic controller. Therefore, a comprehensive framework to generate a data set, train the neural network, and produce hardware configuration for the Spiking Neural Network (SNN), a brain-inspired computing paradigm, is also presented in this paper. Moreover, Spike-MCryptCores integrates the hardware SNN inside its architecture to support low-cost and low-latency adaptations. The results show that implemented SNN controller occupies only 2.3 % of the overall area cost while providing the ability to reduce power consumption significantly. The lightweight SNN controller model is trained and tested with up to 95 % accuracy. The maximum difference between the predicted number of cores and the ideal one from the label is one unit only. Under 24 test scenarios, a SNN controller with clock-gating helps Spike-MCryptCores reducing the power consumption by 48.6 % on the average; by 67 % for the best-case scenario, and by 39 % for the worst-case scenario.

Neuron-based back-calculation anti-windup strategy: first results

Back-calculation is an anti-windup strategy that consists in adding a feedback path to act on the integrator of a proportional-integral-derivative (PID) controller when the actuator causes a saturation. Neuromorphic control (NC) uses neuron models to encode the control signal into pulses. This paper is a first attempt towards using basis of NC as an anti-windup method in a PID controller. More precisely, a neuron is utilized to encode the saturation error (namely, the difference between the controller and the actuator outputs) into pulses, causing that the back-calculation works in lapses according to error intensity. A classical control problem with saturation is given as illustrative example of application of the proposed strategy.

Ion-Mediated Recombination Dynamics in Perovskite-Based Memory Light-Emitting Diodes for Neuromorphic Control Systems.

Neuromorphic devices can help perform memory-heavy tasks more efficiently due to the co-localization of memory and computing. In biological systems, fast dynamics are necessary for rapid communication, while slow dynamics aid in the amplification of signals over noise and regulatory processes such as adaptation- such dual dynamics are key for neuromorphic control systems. Halide perovskites exhibit much more complex phenomena than conventional semiconductors due to their coupled ionic, electronic, and optical properties which result in modulatable drift, diffusion of ions, carriers, and radiative recombination dynamics. This is exploited to engineer a dual-emitter tandem device with the requisite dual slow-fast dynamics. Here, a perovskite-organic tandem light-emitting diode (LED) capable of modulating its emission spectrum and intensity owing to the ion-mediated recombination zone modulation between the green-emitting quasi-2D perovskite layer and the red-emitting organic layer is introduced. Frequency-dependent response and high dynamic range memory of emission intensity and spectra in a LED are demonstrated. Utilizing the emissive read-out, image contrast enhancement as a neuromorphic pre-processing step to improve pattern recognition capabilities is illustrated. As proof of concept using the device's slow-fast dynamics, an inhibition of the return mechanism is physically emulated.

Prevention of emergency situations in complex technical systems using a neuromorphic approach

The paper proposes a scheme of emergency prevention system based on neuromorphic approach. The system includes a prediction unit that implements a mathematical model of the cerebellum predictive functions, and an alarm unit that implements the pain sensation model, proposed by the authors earlier. As a basic element of the proposed system the Compartmental Spiking Neuron Model (CSNM) was used, capable of learning from a small number of examples. The use of neuromorphic approach allows to overcome the limitations associated with the formalizing complexity of the systems being diagnosed and the low availability of data for modeling the processes occurring in them. The overcoming these limitations is possible due to the possibility of learning from a small number of examples and the absence of the need to model the system being diagnosed itself. The paper also presents the results of testing of the proposed scheme, which was carried out on a computer model using synthetic data. The results of the testing showed the fundamental applicability of the proposed scheme in neuromorphic control systems.

Memristive devices with short-term and long-term memory behaviors for processing temporal information

Memristors based on 2D semiconductors such as MoS2 and its derivative materials exhibit analog switching behaviors capable of emulating some synaptic functions, including short-term plasticity, long-term potentiation, and spike-time-dependent-plasticity. Additional investigation is needed to realize reliable control of such synaptic behaviors for practical device implementation. To meet this scientific need, we fabricated MoS2-based memristors and studied their paired-pulse facilitation (PPF) and long-term memory characteristics under different pulse programming settings. This research has provided a guideline for identifying the programming settings for different neuromorphic processes. For example, a specific setting resulting in PPF > 30% and long-term conductance change < 20% has been identified to be suited for processing real-time temporal information. Furthermore, this research also indicates that the MoS2 memristor keeps having an almost constant relative change in conductance but greatly enhanced drive current level under laser illumination. This behavior can enable an easy integration of such memristive devices with state-of-the-art controller circuits for practice neuromorphic control applications.

Autonomous driving controllers with neuromorphic spiking neural networks.

Autonomous driving is one of the hallmarks of artificial intelligence. Neuromorphic (brain-inspired) control is posed to significantly contribute to autonomous behavior by leveraging spiking neural networks-based energy-efficient computational frameworks. In this work, we have explored neuromorphic implementations of four prominent controllers for autonomous driving: pure-pursuit, Stanley, PID, and MPC, using a physics-aware simulation framework. We extensively evaluated these models with various intrinsic parameters and compared their performance with conventional CPU-based implementations. While being neural approximations, we show that neuromorphic models can perform competitively with their conventional counterparts. We provide guidelines for building neuromorphic architectures for control and describe the importance of their underlying tuning parameters and neuronal resources. Our results show that most models would converge to their optimal performances with merely 100-1,000 neurons. They also highlight the importance of hybrid conventional and neuromorphic designs, as was suggested here with the MPC controller. This study also highlights the limitations of neuromorphic implementations, particularly at higher (> 15 m/s) speeds where they tend to degrade faster than in conventional designs.

A perspective on the neuromorphic control of legged locomotion in past, present, and future insect-like robots

This article is a historical perspective on how the study of the neuromechanics of insects and other arthropods has inspired the construction, and especially the control, of hexapod robots. Many hexapod robots’ control systems share common features, including: 1. Direction of motor output of each joint (i.e. to flex or extend) in the leg is gated by an oscillatory or bistable gating mechanism; 2. The relative phasing between each joint is influenced by proprioceptive feedback from the periphery (e.g. joint angles, leg load) or central connections between joint controllers; and 3. Behavior can be directed (e.g. transition from walking along a straight path to walking along a curve) via low-dimensional, broadly-acting descending inputs to the network. These distributed control schemes are inspired by, and in some robots, closely mimic the organization of the nervous systems of insects, the natural hexapods, as well as crustaceans. Nearly a century of research has revealed organizational principles such as central pattern generators, the role of proprioceptive feedback in control, and command neurons. These concepts have inspired the control systems of hexapod robots in the past, in which these structures were applied to robot controllers with neuromorphic (i.e. distributed) organization, but not neuromorphic computational units (i.e. neurons) or computational hardware (i.e. hardware-accelerated neurons). Presently, several hexapod robots are controlled with neuromorphic computational units with or without neuromorphic organization, almost always without neuromorphic hardware. In the near future, we expect to see hexapod robots whose controllers include neuromorphic organization, computational units, and hardware. Such robots may exhibit the full mobility of their insect counterparts thanks to a ‘biology-first’ approach to controller design. This perspective article is not a comprehensive review of the neuroscientific literature but is meant to give those with engineering backgrounds a gentle introduction into the neuroscientific principles that underlie models and inspire neuromorphic robot controllers. A historical summary of hexapod robots whose control systems and behaviors use neuromorphic elements is provided. Robots whose controllers closely model animals and may be used to generate concrete hypotheses for future animal experiments are of particular interest to the authors. The authors hope that by highlighting the decades of experimental research that has led to today’s accepted organization principles of arthropod nervous systems, engineers may better understand these systems and more fully apply biological details in their robots. To assist the interested reader, deeper reviews of particular topics from biology are suggested throughout.

Bioinspired smooth neuromorphic control for robotic arms

Beyond providing accurate movements, achieving smooth motion trajectories is a long-standing goal of robotics control theory for arms aiming to replicate natural human movements. Drawing inspiration from biological agents, whose reaching control networks effortlessly give rise to smooth and precise movements, can simplify these control objectives for robot arms. Neuromorphic processors, which mimic the brain’s computational principles, are an ideal platform to approximate the accuracy and smoothness of biological controllers while maximizing their energy efficiency and robustness. However, the incompatibility of conventional control methods with neuromorphic hardware limits the computational efficiency and explainability of their existing adaptations. In contrast, the neuronal subnetworks underlying smooth and accurate reaching movements are effective, minimal, and inherently compatible with neuromorphic hardware. In this work, we emulate these networks with a biologically realistic spiking neural network for motor control on neuromorphic hardware. The proposed controller incorporates experimentally-identified short-term synaptic plasticity and specialized neurons that regulate sensory feedback gain to provide smooth and accurate joint control across a wide motion range. Concurrently, it preserves the minimal complexity of its biological counterpart and is directly deployable on Intel’s neuromorphic processor. Using the joint controller as a building block and inspired by joint coordination in human arms, we scaled up this approach to control real-world robot arms. The trajectories and smooth, bell-shaped velocity profiles of the resulting motions resembled those of humans, verifying the biological relevance of the controller. Notably, the method achieved state-of-the-art control performance while decreasing the motion jerk by 19% to improve motion smoothness. Overall, this work suggests that control solutions inspired by experimentally identified neuronal architectures can provide effective, neuromorphic-controlled robots.

Neuromorphic control of a simulated 7-DOF arm using Loihi

In this paper, we present a fully spiking neural network running on Intel’s Loihi chip for operational space control of a simulated 7-DOF arm. Our approach uniquely combines neural engineering and deep learning methods to successfully implement position and orientation control of the end effector. The development process involved four stages: (1) Designing a node-based network architecture implementing an analytical solution; (2) developing rate neuron networks to replace the nodes; (3) retraining the network to handle spiking neurons and temporal dynamics; and finally (4) adapting the network for the specific hardware constraints of the Loihi. We benchmark the controller on a center-out reaching task, using the deviation of the end effector from the ideal trajectory as our evaluation metric. The RMSE of the final neuromorphic controller running on Loihi is only slightly worse than the analytic solution, with 4.13% more deviation from the ideal trajectory, and uses two orders of magnitude less energy per inference than standard hardware solutions. While qualitative discrepancies remain, we find these results support both our approach and the potential of neuromorphic controllers. To the best of our knowledge, this work represents the most advanced neuromorphic implementation of neurorobotics developed to date.

Fractional Neuromorphic Controller for a Servomotor: Analog Implementation*

Taking the benefits of each of them individually, this paper investigates the combination of bioinspired pulse-based control, also known as neuromorphic, together with fractional order control, which will be referred to as fractional neuromorphic control (FNC). The proposed strategy essentially consists of a fractional order controller whose output is connected to a neuron to generate pulse-like control signals, i.e., the control signal is encoded into the distance between pulses. An analog implementation of FNC is presented, providing a set of recommendations to address its design. To illustrate the control possibilities that offer this novel strategy, the velocity control of a servomotor is given as an example of application.

Hybrid implementation of neuromorphic system for robot control

Abstract The article proposes a new approach to the implementation of a neuromorphic system based on a hybrid model of a neuron, made in a neuro-fuzzy basis. The closest analogue, the Neo-fuzzy neuron model, is described. A neurological cluster model, previously developed with the participation of the authors, was chosen as a prototype for constructing a hybrid model. The basic procedures for processing and teaching this model to map a given function, as well as additional procedures for coding pulse inputs and synthesis of pulse outputs are presented. Using these procedures, a new hybrid model of a spike neuron has been implemented, which reflects the function of the traditional model of Izhikevich's neuron. It is shown that the hybrid model requires less computing resources. On the basis of the new version of the hybrid model, a spike neuromorphic network was developed and tested to control the movements of a mobile robot along a given trajectory. It is assumed that, using a hybrid version of the neuron model, it is easier to implement a hardware version of spike neuromorphic networks than using neuron models based on differential equations. On the basis of such hardware-implemented networks, it is possible to create effective neuromorphic control systems for mobile robots. Key words Neuromorphic system, spike neuromorphic network, hybrid neuron model, neuron-fuzzy basis, mobile robot control. Acknowledgements This work was done as the part of the state task of the Ministry of Education and Science of Russia No. 075-00913-21-01 «Development and study of new architectures of reconfigurable growing neural networks, methods and algorithms for their learning».

From Theoretical Network to Bedside: Translational Application of Brain-Inspired Computing in Clinical Medicine

Advances in the brain-inspired computing space are growing at a rapid rate, and many of these emerging strategies are in the field of neuromorphic control, robotics, and sensor development, just to name a few. These innovations are disruptive in their own right and have numerous, multi-dimensional medical applications within precision medicine, telematics, device development, and informed clinical decision making. For this discussion, I will define brain-inspired computing in the scope of simulating the architecture of the brain and discuss the realization of integrating hardware and other technologies with the applications of medicine, along with the considerations for the regulatory pathway for approval and evaluating the risk/consequences of failure modes. This perspective is a call for continued discussion of the development of a pathway for translating these technologies into medical treatment and diagnostic strategies. The aim is to align with global regulatory bodies and ensure that regulation does not limit the capacity of these emerging innovations while ensuring patient safety and clinical efficacy. It is my perspective that it is and will continue to be critical that these technologies are correctly perceived and understood in the lens of multiple disciplines in order to reach their full potential for medical applications.

Application of a Compartmental Spiking Neuron Model with Structural Adaptation for Solving Classification Problems

The problem of classification using a compartmental spiking neuron model is considered. The state of the art of spiking neural networks analysis is carried out. It is concluded that there are very few works on the study of compartmental neuron models. The choice of a compartmental spiking model is justified as a neuron model for this work. A brief description of such a model is given, and its main features are noted in terms of the possibility of its structural reconfiguration. The method of structural adaptation of the model to the input spike pattern is described. The general scheme of the compartmental spiking neurons’ organization into a network for solving the classification problem is given. The time-to-first-spike method is chosen for encoding numerical information into spike patterns, and a formula is given for calculating the delays of individual signals in the spike pattern when encoding information. Brief results of experiments on solving the classification problem on publicly available data sets (Iris, MNIST) are presented. The conclusion is made about the comparability of the obtained results with the existing classical methods. In addition, a detailed step-by-step description of experiments to determine the state of an autonomous uninhabited underwater vehicle is provided. Estimates of computational costs for solving the classification problem using a compartmental spiking neuron model are given. The conclusion is made about the prospects of using spiking compartmental models of a neuron to increase the bio-plausibility of the implementation of behavioral functions in neuromorphic control systems. Further promising directions for the development of neuromorphic systems based on the compartmental spiking neuron model are considered.

Pace Running of a Quadruped Robot Driven by Pneumatic Muscle Actuators: An Experimental Study

Our goal is to design a neuromorphic locomotion controller for a prospective bioinspired quadruped robot driven by artificial muscle actuators. In this paper, we focus on achieving a running gait called a pace, in which the ipsilateral pairs of legs move in phase, while the two pairs together move out of phase, by a quadruped robot with realistic legs driven by pneumatic muscle actuators. The robot is controlled by weakly coupled two-level central pattern generators to generate a pace gait with leg loading feedback. Each leg is moved through four sequential phases like an animal, i.e., touch-down, stance, lift-off, and swing phases. We find that leg loading feedback to the central pattern generator can contribute to stabilizing pace running with an appropriate cycle autonomously determined by synchronizing each leg’s oscillation with the roll body oscillation without a human specifying the cycle. The experimental results conclude that our proposed neuromorphic controller is beneficial for achieving pace running by a muscle-driven quadruped robot.

Evolutionary vs imitation learning for neuromorphic control at the edge*

Abstract Neuromorphic computing offers the opportunity to implement extremely low power artificial intelligence at the edge. Control applications, such as autonomous vehicles and robotics, are also of great interest for neuromorphic systems at the edge. It is not clear, however, what the best neuromorphic training approaches are for control applications at the edge. In this work, we implement and compare the performance of evolutionary optimization and imitation learning approaches on an autonomous race car control task using an edge neuromorphic implementation. We show that the evolutionary approaches tend to achieve better performing smaller network sizes that are well-suited to edge deployment, but they also take significantly longer to train. We also describe a workflow to allow for future algorithmic comparisons for neuromorphic hardware on control applications at the edge.

Neuromorphic Control [About This Issue

This issue of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">IEEE Control Systems</i> includes two features and one “Focus on Education” article. The feature “Neuromorphic Control” draws from the Ph.D. dissertation of my coauthor and former student Luka Ribar. Neuromorphic engineering is a rapidly developing field that takes inspiration from the biological organization of neural systems to develop a novel technology for computing, sensing, and actuating. The unique properties of such systems call for new signal processing and control paradigms. The article discusses the mixed-feedback organization of excitable neuronal systems, which consist of interlocked positive and negative feedback loops acting in distinct time scales. The principles of biological neuromodulation suggest a methodology for designing and controlling mixed-feedback systems neuromorphically.

LiDAR-driven spiking neural network for collision avoidance in autonomous driving

Facilitated by advances in real-time sensing, low and high-level control, and machine learning, autonomous vehicles draw ever-increasing attention from many branches of knowledge. Neuromorphic (brain-inspired) implementation of robotic control has been shown to outperform conventional control paradigms in terms of energy efficiency, robustness to perturbations, and adaptation to varying conditions. Here we propose LiDAR-driven neuromorphic control of both vehicle's speed and steering. We evaluated and compared neuromorphic PID control and online learning for autonomous vehicle control in static and dynamic environments, finally suggesting proportional learning as a preferred control scheme. We employed biologically plausible basal-ganglia and thalamus neural models for steering and collision-avoidance, finally extending them to support a null controller and a target-reaching optimization, significantly increasing performance.

BEHAVIORAL FUNCTIONS IMPLEMENTATION ON SPIKING NEURAL NETWORKS

The question of behavioral functions modeling of animals (in particular, the modeling and implementation of the conditioned reflex) is considered. The analysis of the current state of neural networks with the possibility of structural reconfiguration is carried out. The modeling is carried out by means of neural networks, which are built on the basis of a compartmental spiking model of a neuron with the possibility of structural adaptation to the input pulse pattern. The compartmental spike model of a neuron is able to change its structure (the size of the cell body, the number and length of dendrites, the number of synapses) depending on the incoming pulse pattern at its inputs. A brief description of the compartmental spiking model of a neuron is given, and its main features are noted in terms of the possibility of its structural reconfiguration. The method of structural adaptation of the compartmental spiking model of the neuron to the input pulse pattern is described. To study the work of the proposed model of a neuron in a network, the choice of a conditioned reflex as a special case of the formation of associative connections is justified as an example. The structural scheme and algorithm of formation of a conditioned reflex with both positive and negative reinforcement are described. The article presents a step-by-step description of experiments on the associative connection’s formation in general and conditioned reflex (both with positive and negative reinforcement), in particular. The conclusion is made about the prospects of using spiking compartmental models of neurons to improve the efficiency of the implementation of behavioral functions in neuromorphic control systems. Further promising directions for the development of neuromorphic systems based on spiking compartmental models of the neuron are considered.

Neuromorphic NEF-Based Inverse Kinematics and PID Control.

Neuromorphic implementation of robotic control has been shown to outperform conventional control paradigms in terms of robustness to perturbations and adaptation to varying conditions. Two main ingredients of robotics are inverse kinematic and Proportional–Integral–Derivative (PID) control. Inverse kinematics is used to compute an appropriate state in a robot's configuration space, given a target position in task space. PID control applies responsive correction signals to a robot's actuators, allowing it to reach its target accurately. The Neural Engineering Framework (NEF) offers a theoretical framework for a neuromorphic encoding of mathematical constructs with spiking neurons for the implementation of functional large-scale neural networks. In this work, we developed NEF-based neuromorphic algorithms for inverse kinematics and PID control, which we used to manipulate 6 degrees of freedom robotic arm. We used online learning for inverse kinematics and signal integration and differentiation for PID, offering high performing and energy-efficient neuromorphic control. Algorithms were evaluated in simulation as well as on Intel's Loihi neuromorphic hardware.

Evolved Neuromorphic Control for High Speed Divergence-Based Landings of MAVs

Flying insects are capable of vision-based navigation in cluttered environments, reliably avoiding obstacles through fast and agile maneuvers, while being very efficient in the processing of visual stimuli. Meanwhile, autonomous micro air vehicles still lag far behind their biological counterparts, displaying inferior performance at a much higher energy consumption. In light of this, we want to mimic flying insects in terms of their processing capabilities, and consequently show the efficiency of this approach in the real world. This letter does so through evolving spiking neural networks for controlling landings of micro air vehicles using optical flow divergence from a downward-looking camera. We demonstrate that the resulting neuromorphic controllers transfer robustly from a highly abstracted simulation to the real world, performing fast and safe landings while keeping network spike rate minimal. Furthermore, we provide insight into the resources required for successfully solving the problem of divergence-based landing, showing that high-resolution control can be learned with only a single spiking neuron. To the best of our knowledge, this work is the first to integrate spiking neural networks in the control loop of a real-world flying robot. Videos of the experiments can be found at https://bit.ly/neuro-controller .