Hardware Counterpart Research Articles

Physical Reservoir Computing Enabled by Solitary Waves and Biologically Inspired Nonlinear Transformation of Input Data

Reservoir computing (RC) systems can efficiently forecast chaotic time series using the nonlinear dynamical properties of an artificial neural network of random connections. The versatility of RC systems has motivated further research on both hardware counterparts of traditional RC algorithms and more-efficient RC-like schemes. Inspired by the nonlinear processes in a living biological brain and using solitary waves excited on the surface of a flowing liquid film, in this paper, we experimentally validated a physical RC system that substitutes the effect of randomness that underpins the operation of the traditional RC algorithm for a nonlinear transformation of input data. Carrying out all operations using a microcontroller with minimal computational power, we demonstrate that the so-designed RC system serves as a technically simple hardware counterpart to the ‘next-generation’ improvement of the traditional RC algorithm.

Intelligent Real-Time Face-Mask Detection System with Hardware Acceleration for COVID-19 Mitigation.

This paper proposes and implements a dedicated hardware accelerated real-time face-mask detection system using deep learning (DL). The proposed face-mask detection model (MaskDetect) was benchmarked on three embedded platforms: Raspberry PI 4B with either Google Coral USB TPU or Intel Neural Compute Stick 2 VPU, and NVIDIA Jetson Nano. The MaskDetect was independently quantised and optimised for each hardware accelerated implementation. An ablation study was carried out on the proposed model and its quantised implementations on the embedded hardware configurations above as a comparison to other popular transfer-learning models, such as VGG16, ResNet-50V2, and InceptionV3, which are compatible with these acceleration hardware platforms. The ablation study revealed that MaskDetect achieved excellent average face-mask detection performance with accuracy above 94% across all embedded platforms except for Coral, which achieved an average accuracy of nearly 90%. With respect to detection performance (accuracy), inference speed (frames per second (FPS)), and product cost, the ablation study revealed that implementation on Jetson Nano is the best choice for real-time face-mask detection. It achieved 94.2% detection accuracy and twice greater FPS when compared to its desktop hardware counterpart.

Adapting a Dehazing System to Haze Conditions by Piece-Wisely Linearizing a Depth Estimator.

Haze is the most frequently encountered weather condition on the road, and it accounts for a considerable number of car crashes occurring every year. Accordingly, image dehazing has garnered strong interest in recent decades. However, although various algorithms have been developed, a robust dehazing method that can operate reliably in different haze conditions is still in great demand. Therefore, this paper presents a method to adapt a dehazing system to various haze conditions. Under this approach, the proposed method discriminates haze conditions based on the haze density estimate. The discrimination result is then leveraged to form a piece-wise linear weight to modify the depth estimator. Consequently, the proposed method can effectively handle arbitrary input images regardless of their haze condition. This paper also presents a corresponding real-time hardware implementation to facilitate the integration into existing embedded systems. Finally, a comparative assessment against benchmark designs demonstrates the efficacy of the proposed dehazing method and its hardware counterpart.

Complexity Measure of Software Composition Framework

This article proposes a pragmatic approach to software composition based on matching criteria by mimicking integrated hardware counterpart. The tangible value of consumer goods as described by Cox instills this logical derivation of the proposed approach. As software gradually matures in component form, various software compositions can be systematically assembled from related existing software. Two application software are composed based on their functionality matching. Their composition complexities are measured by function point. The total effort unveils a noteworthy finding that high complexity software demands larger effort to deploy. Thus, the proposed software composition framework not only offers the freedom of development mandates to the development team, but also broadens the horizon of cost and project evaluations by arriving at the proper mix of constituent components in the software product.

FINE–GRAINED ANALYSIS AND PROFILING OF SOFTWARE BUGS TO FACILITATE WASTE IDENTIFICATION AND ITS MINIMIZATION

Software defects or bugs are among the primary causes of software development overrunning time schedules and budget costs. They are also the major cause of ‘waste’ in software development, which roughly translates to time, effort, and money spent on unproductive aspects of software. Despite several developments in software engineering and improvements in the software development process, their effects in minimizing the waste in software development process have not been remarkable in comparison with the hardware counterpart of complex chip design. This paper proposes a fine–grained approach to the analysis of the root causes of software defects (bugs) in an effort to better quantify the components of waste and its subsequent minimization. It also proposes the use of bugs profile for the allocation of resources to tackle bugs with minimal wasted resources.

Efficient Architecture For Island Genetic Algorithm in Reconfigurable Hardware

Abstract A novel VLSI architecture for an island genetic algorithm (GA) is presented in this paper. The island GA is based on steady-state GA for reducing the hardware resources consumption. Alook-up table based fast string migration architecture is proposed for lowering the computational overhead while enhancing the performance for the island GA. As compared with its single-island GA hardware counterpart, the proposed architecture attains superior performance with less computation time subject to the same total population size. In addition, the proposed architecture has significantly lower computational time as compared with its software counterparts running on cluster computer with multithreading for GA-based optimization.

Evolving neuromorphic flight control for a flapping‐wing mechanical insect

PurposeThe purpose of this paper is to present an approach to employ evolvable hardware concepts, to effectively construct flapping‐wing mechanism controllers for micro robots, with the evolved dynamically complex controllers embedded in a, physically realizable, micro‐scale reconfigurable substrate.Design/methodology/approachIn this paper, a continuous time recurrent neural network (CTRNN)‐evolvable hardware (a neuromorphic variant of evolvable hardware) framework and methodologies are employed in the process of designing the evolution experiments. CTRNN is selected as the neuromorphic reconfigurable substrate with most efficient Minipop Evolutionary Algorithm, configured to drive the evolution process. The uniqueness of the reconfigurable CTRNN substrate preferred for this study is perceived from its universal dynamics approximation capabilities and prospective to realize the same in small area and low power chips, the properties which are very much a basic requirement for flapping‐wing based micro robot control. A simulated micro mechanical flapping insect model is employed to conduct the feasibility study of evolving neuromorphic controllers using the above‐mentioned methodology.FindingsIt has been demonstrated that the presented neuromorphic evolvable hardware approach can be effectively used to evolve controllers, to produce various flight dynamics like cruising, steering, and altitude gain in a simulated micro mechanical insect. Moreover, an appropriate feasibility is presented, to realize the evolved controllers in small area and lower power chips, with available fabrication techniques and as well as utilizing the complex dynamics nature of CTRNNs to encompass various controls ability in a architecturally static hardware circuit, which are more pertinent to meet the constraints of micro robot construction and control.Originality/valueThe proposed neuromorphic evolvable hardware approach along with its modules intact (CTRNNs and Minipop) can provide a general mechanism to construct/evolve dynamically complex and optimal controllers for flapping‐wing mechanism based micro robots for various environments with least human intervention. Further, the evolved neuromorphic controllers in simulation study can be successfully transferred to its hardware counterpart without sacrificing its anticipated functionality and realized within a predictable area and power ranges.

Interpolator for a Computer Numerical Control System

A software interpolator which is comprised of linear and circular interpolations is compared with its hardware counterpart and with other circular interpolation methods. The software interpolator and the feed-rate control are contained in the numerical control (NC) program of a computer numerical control (CNC) system and enable a contouring control of the machine tool in any required feed-rate.

A Computer Network for Improved Control

The bases of a practical scheme for integrated control of modern industrial complexes is decomposition and hierarchical organization. Such control scheme requires a new look at the hardware which is needed for its implementation. The proper candidate for such job are not large computers but rather networks of small computers, LSI processor chips and memory chips. A network consisting of such processors and data transmission channels with the processors and the processing loads connected through inteligent interfacing nodes to the data transmission channel is most adequate. This form of realization is the hardware counterpart of the hierachical organization of control and management process.