The observations of linear and nonlinear physical processes are subject to random errors, which can be represented by a wide variety of probability distributions. In contrast, most estimation and inference techniques rely on a Gaussian assumption, which may limit our ability to make model-based predictions. There is a need for data assimilation methods that can capture and leverage the higher moments of these physical processes for state estimation and forecasting. In this paper, we develop the generalized unscented transform (GenUT), which uses a minimal number of sample points to accurately capture elements of the higher moments of most probability distributions. Constraints can be analytically enforced on the sample points while guaranteeing at least second-order accuracy. The GenUT is widely applicable to non-Gaussian distributions, which can substantially improve the assimilation of observations of nonlinear physics, such as the modeling of infectious diseases.

A complex system is made up of multiple related and coupled subsystems. Each subsystem has its own set of multiple constraints and objectives, and is commonly found in real-world applications. Biogeography-based complex system optimization (BBO/Complex) is a population-based evolutionary intelligence paradigm that has been developed to solve complex system problems. The paper first presents the framework of the proposed method that consists of within-subsystem migration, cross-subsystem migration, and mutation operators. Then a theoretical model of the proposed method is derived using Markov chain and the probability transition matrix of each operator, along with the generalized multinomial theorem. The model provides exact mathematical formulas that predict the limiting probabilities of optimization performance for each possible population in the proposed method. Additionally, simulation results of two sample problems, which include multiple subsystems and multiple objectives, confirm the Markov model derived. Finally, the performance of the proposed method is investigated on an application of a battery charging system, and numerical simulation shows that the proposed method is a promising optimization method for the studied complex system.

Ground reaction force (GRF) is a potentially useful control input for powered lower limb prostheses but accurate GRF measurement in real-time is challenging. The objective of this work is to evaluate the ability to estimate GRF from a minimal set of kinematic inputs (knee and ankle angle and angular velocities) during walking. Three artificial neural networks (ANNs) are evaluated for this purpose: nonlinear autoregression with exogenous input (NARX), delayed discrete recurrent neural network (DDRNN), and a self-organizing map with feedforward neural network (SOM-FFNN). Specifically, our work focuses on investigating the impact of ANN architecture and training/learning algorithms on the predication accuracy of GRF. First, ANN performance in open loop GRF estimation is investigated using treadmill walking data in a healthy participant at speeds from 0.79 to 1.9 m/s. Next, the effect of ANN estimated GRF is evaluated in a simulation of closed-loop powered prosthesis control with three levels of measurement noise. The results show that all ANNs are able to estimate GRF in open-loop with relatively low RMS, although SOM-FFNN performed the best with an average RMS of 4.85 N across all gait speeds. SOM-FFNN also showed the most robust performance in estimating GRF for trajectory tracking in closed-loop control, providing impetus for its further investigation in control of powered prostheses.

This work proposes the design of an optimization method for high-power LED luminaires with the introduction of new evaluation metrics. A luminaire geometry computational method is deployed to conduct thermal and optical analysis. This current effort novels by designing a tool that enables the analysis of uniformity for individual luminaire over the target plane in accordance with international regulatory standards. Additionally, adequate thermal management is conducted to guarantee nominal operation standard values determined by LED vendors. The results of this optimization method present luminaire models with different geometries that allow the stabilization of the temperature within the safety and uniform illuminance distribution thresholds. The resulting solution proposes the design of a 2×2 HP-LED rectangular luminaire. During simulations, the temperature of the LED reaches a maximum value of 73.9∘ C in a steady state with a uniform index of 0.228 for its individual luminaire. The overall uniform index identified for two separate and adjacent luminaire points in a pedestrian walk is 0.5413 with a minimal illuminance of 36.95 lx, maximum illuminance of 93.65 lx and average illuminance of 68.27 lx. Overall, we conclude that the currently adopted metric, which takes into consideration only the ratio between the minimum and the average illuminance, is not efficient and it cannot distinguish different luminaire geometry standards according to their uniform illuminance distribution. The metric proposed and designed in this work is capable of evaluating illuminance and thermal threshold criteria, as well as classifying different sorts of luminaries.

Individuals with an above-knee (AK) amputation typically use passive prostheses, whether reactive (microprocessor) or purely mechanical. Though sufficient for walking, these solutions lack the positive power generation observed in able-bodied individuals. Active (powered) prostheses can provide positive power but suffer complex control and limited energy storage capacities. These shortcomings motivate the development of an active prosthesis implementing a novel impedance controller design with energy regeneration. The controller requires only five tuning parameters that are intuitive to adjust in contrast to the current standard—finite state machine impedance scheduling of up to 45 gains. This simplification is uniquely achieved by modulating knee joint impedance by axial shank force. Furthermore, the proposed control approach introduces analytical guidance for impedance tuning to purposely integrate energy regeneration; specifically, a precise amount of negative damping is injected into the joint. A pilot study conducted with a volunteer with an AK amputation walking at three distinct speeds and at continually self-selected varying speeds demonstrated the adaptability of the controller to changes in speed. Self-powered operation was attained for all trials despite low mechanical component efficiencies. These early results suggest the efficacy of simplifying impedance control tuning and fusing control and energy regeneration in transfemoral prostheses.

During development of power Integrated Circuits (IC), several iterations between the design and test/ measurement steps are performed. Computer-aided engineering significantly shortens the product development process because the numerical simulations can identify and remediate most deficiencies during the design stage. The recent IC manufacturing technologies lead to ca. $10^{4}$ -order scale separation between transistor cell details and the device active area, resulting in very complex IC models. For the IC complexity to be overcome, advanced multi-scale analysis methods are required to perform accurate simulations in a decent time (order of hours). This paper proposes an advanced and enhanced multi-scale simulation method for the thermo-mechanical analysis of power ICs. The computational IC structure is automatically generated from a Cadence layout and partitioned into far-field and homogenized regions - the macro-model. Detailed localized micro-scale sub-models are assigned to limited portions of the homogenized region. The two-way simulated data transfer between the homogenized macro-model and the micro sub-models is one multi-scale approach novelty proposed in this paper. The method is validated on a real test chip structure presented in literature. The proposed multi-scale approach in conjunction with the two-way macro-micro data transfer lead to similar accuracy in the prediction of defect location, yet with significant simulation time - and computational resource reduction (CPU time and RAM usage reduced by almost 80% and 60% respectively) compared to the method used as reference.

Complex system optimization is an emerging research topic in the field of evolutionary computation, whose goal is to handle complex systems with multiple coupled subsystems, each including multiple objectives and multiple constraints in real-world applications. This paper proposes a multi-system genetic algorithm (MSGA), stemming from implicit parallelism in population-based search algorithms, to solve multiple coupled subsystems simultaneously in a complex system. The proposed MSGA is composed of within-subsystem evolution and cross-subsystem migration operators. The objective of the former is to optimize each subsystem by appropriate search strategies, and the objective of the latter is to exchange information between multiple subsystems by migration, which is based on the similarity probability of objectives and constraints, and the intersection probability of solutions in different subsystems. During migration across subsystems, three statistical approaches of measuring similarity and three metrics of solution intersection in information theory are used to calculate these probabilities. Performance is tested on a set of multi-subsystem benchmark functions, and the simulation results show that cross-subsystem migration plays the key role for the performance of MSGA. Furthermore, the proposed MSGA is compared with other competitive algorithms, and results show that it is a promising multi-system optimization algorithm. In summary, the contribution of this paper is the introduction of multi-system optimization to the EA community.

This paper develops a robust regressor-free controller for n -link robots with state constraints. We use the function approximation technique to represent the uncertain robot dynamics. Our controller, which uses a state-dependent barrier Lyapunov function, prevents the states from violating their constraints by guaranteeing uniform ultimate boundedness of the closed-loop dynamics via a fixed control structure and a continuous switching law. In contrast to the drawbacks of controllers developed via the widely used error-dependent barrier Lyapunov functions, our controller overcomes the need to impose constraints on the initial tracking errors and the reference trajectories. When the reference trajectories violate the state constraints, our controller is capable of maintaining states within their constraints. We show that it is easy to switch between the constrained and unconstrained versions of our controller. Our controller requires few tuning parameters and is robust to wide ranges of uncertainties. Simulation and real-time experimental tests validate the practicality of our proposed controller.

This paper presents and experimentally implements three different adaptive and robust adaptive controllers as the first steps toward using model-based controllers for transfemoral prostheses. The goal of this paper is to translate these control methods to the robotic domain, from bipedal robotic walking to prosthesis walking, including a rigorous stability analysis. The human/prosthesis system is first modeled as a two-domain hybrid asymmetric system. An optimization problem is formulated to obtain a stable human-like gait. The proposed controllers are then developed for the combined human/prosthesis model and the optimized reference gait. The stability of all three controllers is proven using the Lyapunov stability theorem, ensuring convergence to the desired gait. The proposed controllers are first verified on a bipedal walking robot as a hybrid human/prosthesis model in simulation. They are then experimentally tested on a treadmill with an able-bodied subject using third iteration of AMBER Prosthetic (AMPRO3), a custom self-contained powered transfemoral prosthesis. Finally, outdoor tests are carried out using AMPRO3 with three test subjects walking on level ground, uphill slopes, and downhill slopes at slope angles of 3° and 8°, to demonstrate walking in different real-world environments.

Objective: Locomotion mode recognition (LMR) enables seamless and natural transitions between low-level control systems in a powered prosthesis. We present a new optimization framework for LMR that eliminates irrelevant or redundant features and measurement signals while still maintaining performance. Methods: We use multi-objective biogeography-based optimization to find a compromise solution between performance and the minimization of feature set size. Experimental data are collected from four transfemoral users walking with a powered knee-ankle prosthesis. We compare the performance of LMR systems trained with the optimal feature subsets and with the full feature set using a deep neural network classifier across six locomotion modes: standing, flat-ground walking, stair up/down, and ramp up/down. Results: Statistical tests indicate that classifier performance using the optimal feature subsets is statistically equal to that using the full feature set. The LMR trained with an optimal subset results in the 1.98% steady-state and 4.09% transitional error rates, while only using approximately 41% and 53% of the available features and sensors, respectively. Conclusion: Results thus indicate the capability of the proposed framework to achieve simultaneously accurate and low-complex LMR systems for transfemoral individuals with powered prostheses. Significance: This framework would potentially lead to less frequent clinical visits needed for sensor replacement and calibrations, which may save health care costs and the prosthesis user's time and energy.

During the COVID-19 pandemic and similar outbreaks in the future, drones can be set up to reduce human interaction for medical supplies delivery, which is crucial in times of pandemic. In this short paper, we introduce the use of two evolutionary algorithms for multi-objective optimization (MOO) and tuning the parameters of the PD controller of a drone to follow the 3D desired path.

The unscented transform uses a weighted set of samples called sigma points to propagate the means and covariances of nonlinear transformations of random variables. However, unscented transforms developed using either the Gaussian assumption or a minimum set of sigma points typically fall short when the random variable is not Gaussian distributed and the nonlinearities are substantial. In this paper, we develop the generalized unscented transform (GenUT), which uses 2n+1 sigma points to accurately capture up to the diagonal components of the skewness and kurtosis tensors of most probability distributions. Constraints can be analytically enforced on the sigma points while guaranteeing at least second-order accuracy. The GenUT uses the same number of sigma points as the original unscented transform while also being applicable to non-Gaussian distributions, including the assimilation of observations in the modeling of infectious diseases such as coronavirus (SARS-CoV-2) causing COVID-19.

The ultimate goal of ridesharing systems is to matchtravelers who do not have a vehicle with those travelers whowant to share their vehicle. A good match can be found amongthose who have similar itineraries and time schedules. In thisway each rider can be served without any delay and also eachdriver can earn as much as possible without having too muchdeviation from their original route. We propose an algorithmthat leverages biogeography-based optimization to solve a multi-objective optimization problem for online ridesharing. It isnecessary to solve the ridesharing problem as a multi-objectiveproblem since there are some important objectives that must beconsidered simultaneously. We test our algorithm by evaluatingperformance on the Beijing ridesharing dataset. The simulationresults indicate that BBO provides competitive performancerelative to state-of-the-art ridesharing optimization algorithms.

Drones are effective for reducing human activity and interactions by performing tasks such as exploring and inspecting new environments, monitoring resources and delivering packages. Drones need a controller to maintain stability and to reach their goal. The most well-known drone controllers are proportional-integral-derivative (PID) and proportional-derivative (PD) controllers. However, the controller parameters need to be tuned and optimized. In this paper, we introduce the use of two evolutionary algorithms, biogeography-based optimization (BBO) and particle swarm optimization (PSO), for multi-objective optimization (MOO) to tune the parameters of the PD controller of a drone. The combination of MOO, BBO, and PSO results in various methods for optimization: vector evaluated BBO and PSO, denoted as VEBBO and VEPSO; and non-dominated sorting BBO and PSO, denoted as NSBBO and NSPSO. The multi-objective cost function is based on tracking errors for the four states of the system. Two criteria for evaluating the Pareto fronts of the optimization methods, normalized hypervolume and relative coverage, are used to compare performance. Results show that NSBBO generally performs better than the other methods.

This work proposes a resilient and adaptive state estimation framework for robots operating in perceptually-degraded environments. The approach, called Adaptive Maximum Correntropy Criterion Kalman Filtering (AMCCKF), is inherently robust to corrupted measurements, such as those containing jumps or general non-Gaussian noise, and is able to modify filter parameters online to improve performance. Two separate methods are developed - the Variational Bayesian AMCCKF (VB-AMCCKF) and Residual AMCCKF (R-AMCCKF) - that modify the process and measurement noise models in addition to the bandwidth of the kernel function used in MCCKF based on the quality of measurements received. The two approaches differ in computational complexity and overall performance which is experimentally analyzed. The method is demonstrated in real experiments on both aerial and ground robots and is part of the solution used by the COSTAR team participating at the DARPA Subterranean Challenge.

This paper introduces an extensive human motion data set for typical activities of daily living. These data are crucial for the design and control of prosthetic devices for transfemoral prosthesis users. This data set was collected from seven individuals, including five individuals with intact limbs and two transfemoral prosthesis users. These data include the following types of movements: (1) walking at three different speeds; (2) walking up and down a 5-degree ramp; (3) stepping up and down; (4) sitting down and standing up. We provide full-body marker trajectories and ground reaction forces (GRFs) as well as joint angles, joint velocities, joint torques, and joint powers. This data set is publicly available at the website referenced in this paper. Data from flexion and extension of the hip, knee, and ankle are presented in this paper. However, the data accompanying this paper (available on the internet) include 46 distinct measurements and can be useful for validating or generating mathematical models to simulate the gait of both transfemoral prosthesis users and individuals with intact legs.

This paper demonstrates the use of hardware-in-the-loop (HIL) simulation to mimic control action scenarios in an active prosthetic leg. The simulation is used to carry out a comparative assessment for model-based controllers such as computed torque, the Slotine and Li approach, and a regressor free approach using adaptive control based on the function approximation technique (FAT). The controller side comprises an embedded single-board computer (SBC) commanding a simulation computer. On the other side, the simulation computer runs a multibody model of a robotic setup that mimics hip motion. The HIL simulation shows that the adaptive FAT controller has better tracking with less control effort compared to the computed torque and Slotine controllers.

Accurate attitude estimation using low-cost sensors is an important capability to enable many robotic applications. In this paper, we present a method based on the concept of correntropy in Kalman filtering to estimate the 3D orientation of a rigid body using a low-cost inertial measurement unit (IMU). We then leverage the proposed attitude estimation framework to develop a LiDAR-Intertial Odometry (LIO) demonstrating improved localization accuracy with respect to traditional methods. This is of particular importance when the robot undergoes high-rate motions that typically exacerbate the issues associated with low-cost sensors. The proposed orientation estimation approach is first validated using the data coming from a low-cost IMU sensor. We further demonstrate the performance of the proposed LIO solution in a simulated robotic cave exploration scenario.

Despite the popularity of drones and their relatively simple operation, the underlying control algorithms can be difficult to design due to the drones' underactuation and highly nonlinear properties. This paper focuses on position and orientation control of drones to address challenges such as path and edge tracking, and disturbance rejection. The adaptive function approximation technique control method is used to control an underactuated and nonlinear drone. The controller utilizes reference attitude signals, that are derived from a proportional derivative (PD) linear feedback control methodology. To avoid analytic expressions for the reference attitude velocities, we employ a continuoustime Kalman filter based on a model of the measurement signal - which is derived by passing the reference attitude position through a low-pass signal differentiator - as a second-order New-tonian system. Stability of the closed loop system is proven using a Lyapunov function. Our design methodology simplifies the control process by requiring only a few tuning variables, while being robust to time-varying and time-invariant uncertainties with unknown variation bounds, and avoids the requirement for the knowledge of the dynamic equation that governs the attitude of the drone. Three different scenarios are simulated and our control method shows better accuracy than the proportional-derivative controller in terms of edge tracking and disturbance rejection.