Distance-aware Approximate Nanophotonic Interconnect

The energy consumption of manycore architectures is dominated by data movement, which calls for energy-efficient and high-bandwidth interconnects. To overcome the bandwidth limitation of electrical interconnects, integrated optics appear as a promising technology. However, it suffers from high power overhead related to low laser efficiency, which calls for the use of techniques and methods to improve its energy costs. Besides, approximate computing is emerging as an efficient method to reduce energy consumption and improve execution speed of embedded computing systems. It relies on allowing accuracy reduction on data at the cost of tolerable application output error. In this context, the work presented in this article exploits both features by defining approximate communications for error-tolerant applications. We propose a method to design realistic and scalable nanophotonic interconnect supporting approximate data transmission and power adaption according to the communication distance to improve the energy efficiency. For this purpose, the data can be sent by mixing low optical power signal and truncation for the Least Significant Bits (LSB) of the floating-point numbers, while the overall power is adapted according to the communication distance. We define two ranges of communications, short and long, which require only four power levels. This reduces area and power overhead to control the laser output power. A transmission model allows estimating the laser power according to the targeted BER and the number of truncated bits, while the optical network interface allows configuring, at runtime, the number of approximated and truncated bits and the laser output powers. We explore the energy efficiency provided by each communication scheme, and we investigate the error resilience of the benchmarks over several approximation and truncation schemes. The simulation results of ApproxBench applications show that, compared to an interconnect involving only robust communications, approximations in the optical transmission led to up to 53% laser power reduction with a limited degradation at the application level with less than 9% of output error. Finally, we show that our solution is scalable and leads to 10% reduction in the total energy consumption, 35× reduction in the laser driver size, and 10× reduction in the laser controller compared to state-of-the-art solution.

Barrier Lyapunov function-based robot control with an augmented neural network approximator

Purpose Considering the complexity of dynamic and friction modeling, this paper aims to develop an adaptive trajectory tracking control scheme for robot manipulators in a universal unmodeled method, avoiding complicated modeling processes. Design/methodology/approach An augmented neural network (NN) constituted of radial basis function neural networks (RBFNNs) and additional sigmoid-jump activation function (SJF) neurons is introduced to approximate complicated dynamics of the system: the RBFNNs estimate the continuous dynamic term and SJF neurons handle the discontinuous friction torques. Moreover, the control algorithm is designed based on Barrier Lyapunov Function (BLF) to constrain output error. Findings Lyapunov stability analysis demonstrates the exponential stability of the closed-loop system and guarantees the tracking errors within predefined boundaries. The introduction of SJFs alleviates the limitation of RBFNNs on discontinuous function approximation. Owing to the fast learning speed of RBFNNs and jump response of SJFs, this modified NN approximator can reconstruct the system model accurately at a low compute cost, and thereby better tracking performance can be obtained. Experiments conducted on a manipulator verify the improvement and superiority of the proposed scheme in tracking performance and uncertainty compensation compared to a standard NN control scheme. Originality/value An enhanced NN approximator constituted of RBFNN and additional SJF neurons is presented which can compensate the continuous dynamic and discontinuous friction simultaneously. This control algorithm has potential usages in high-performance robots with unknown dynamic and variable friction. Furthermore, it is the first time to combine the augmented NN approximator with BLF. After more exact model compensation, a smaller tracking error is realized and a more stringent constraint of output error can be implemented. The proposed control scheme is applicable to some constraint occasion like an exoskeleton and surgical robot.

ARX, ARMAX, Box-Jenkins, Output-Error, and Hammerstein Models for Modeling Intelligent Pneumatic Actuator (IPA) System

A pneumatic actuator is highly nonlinear, which makes the precise position control of this actuator difficult to achieve. In order to achieve precise control, selecting a suitable model structure is a prerequisite before control estimation. This selection of the model structure is based upon an understanding of the physical systems. In this paper, the black-box model is chosen as a system identification model for modeling position control of an Intelligent Pneumatic Actuator (IPA) system and a variety of parametric model structures. The parametric model structure, such as ARX, ARMAX, Box-Jenkins, output-error structures, and Hammerstein available in the black-box model, is used to assist in modeling the IPA system. The results indicate that Hammerstein had the best performance for modeling position control of the IPA system with the best fit 94.95. Also, the results show that ARX, ARMAX, Box-Jenkins, and output-error structures had best fit more than 90%.

Aerodynamic coefficients modeling using Levenberg–Marquardt algorithm and network

Purpose The purpose of this paper is to propose a new aerodynamic parameter estimation methodology based on neural network and output error method, while the output error method is improved based on particle swarm algorithm. Design/methodology/approach Firstly, the algorithm approximates the dynamic characteristics of aircraft based on feedforward neural network. Neural network is trained by extreme learning machine, and the trained network can predict the aircraft response at (k + 1)th instant given the measured flight data at kth instant. Secondly, particle swarm optimization is used to enhance the convergence of Levenberg–Marquardt (LM) algorithm, and the improved LM method is used to substitute for the Gauss Newton algorithm in output error method. Finally, the trained neural network is combined with the improved output error method to estimate aerodynamic derivatives. Findings Neither depending on the initial guess of the parameters to be estimated nor requiring numerical integration of the aircraft motion equation, the proposed algorithm can be used for unstable aircraft and is successfully applied to extract aerodynamic derivatives from both simulated and real flight data. Research limitations/implications The proposed method requires iterative calculation and can only identify parameters offline. Practical implications The proposed method is successfully applied to estimate aircraft aerodynamic parameters and can also be used as a new algorithm for other optimization problems. Originality/value In this study, the output error method is improved to reduce the dependence on the initial value of parameters and expand its application scope. It is applied in aircraft aerodynamic parameter identification together with neural network.