Optimizing Human-Robot Teaming Performance through Q-Learning-Based Task Load Adjustment and Physiological Data Analysis.

The transition to Industry 4.0 and 5.0 underscores the need for integrating humans into manufacturing processes, shifting the focus towards customization and personalization rather than traditional mass production. However, human performance during task execution may vary. To ensure high human-robot teaming (HRT) performance, it is crucial to predict performance without negatively affecting task execution. Therefore, to predict performance indirectly, significant factors affecting human performance, such as engagement and task load (i.e., amount of cognitive, physical, and/or sensory resources required to perform a particular task), must be considered. Hence, we propose a framework to predict and maximize the HRT performance. For the prediction of task performance during the development phase, our methodology employs features extracted from physiological data as inputs. The labels for these predictions-categorized as accurate performance or inaccurate performance due to high/low task load-are meticulously crafted using a combination of the NASA TLX questionnaire, records of human performance in quality control tasks, and the application of Q-Learning to derive task-specific weights for the task load indices. This structured approach enables the deployment of our model to exclusively rely on physiological data for predicting performance, thereby achieving an accuracy rate of 95.45% in forecasting HRT performance. To maintain optimized HRT performance, this study further introduces a method of dynamically adjusting the robot's speed in the case of low performance. This strategic adjustment is designed to effectively balance the task load, thereby enhancing the efficiency of human-robot collaboration.

A comparative study of machine learning methods for classifying ERP scalp distribution.

Objective. Machine learning (ML) methods are used in different fields for classification and regression purposes with different applications. These methods are also used with various non-invasive brain signals, including Electroencephalography (EEG) signals to detect some patterns in the brain signals. ML methods are considered critical tools for EEG analysis since could overcome some of the limitations in the traditional methods of EEG analysis such as Event-related potentials (ERPs) analysis. The goal of this paper was to apply ML classification methods on ERP scalp distribution to investigate the performance of these methods in identifying numerical information carried in different finger-numeral configurations (FNCs). FNCs in their three forms of montring, counting, and non-canonical counting are used for communication, counting, and doing arithmetic across the world between children and even adults. Studies have shown the relationship between perceptual and semantic processing of FNCs, and neural differences in visually identifying different types of FNCs.Approach.A publicly available 32-channel EEG dataset recorded for 38 participants while they were shown a picture of an FNC (i.e., three categories and four numbers of 1,2,3, and 4) was used. EEG data were pre-processed and ERP scalp distribution of different FNCs was classified across time by six ML methods, including support vector machine, linear discriminant analysis, naïve Bayes, decision tree, K-nearest neighbor, and neural network. The classification was conducted in two conditions: classifying all FNCs together (i.e., 12 classes) and classifying FNCs of each category separately (i.e., 4 classes).Results.The support vector machine had the highest classification accuracy for both conditions. For classifying all FNCs together, the K-nearest neighbor was the next in line; however, the neural network could retrieve numerical information from the FNCs for category-specific classification.Significance.The significance of this study is in exploring the application of multiple ML methods in recognizing numerical information contained in ERP scalp distribution of different finger-numeral configurations.

Experiment protocols for brain-body imaging of locomotion: A systematic review.

Introduction: Human locomotion is affected by several factors, such as growth and aging, health conditions, and physical activity levels for maintaining overall health and well-being. Notably, impaired locomotion is a prevalent cause of disability, significantly impacting the quality of life of individuals. The uniqueness and high prevalence of human locomotion have led to a surge of research to develop experimental protocols for studying the brain substrates, muscle responses, and motion signatures associated with locomotion. However, from a technical perspective, reproducing locomotion experiments has been challenging due to the lack of standardized protocols and benchmarking tools, which impairs the evaluation of research quality and the validation of previous findings. Methods: This paper addresses the challenges by conducting a systematic review of existing neuroimaging studies on human locomotion, focusing on the settings of experimental protocols, such as locomotion intensity, duration, distance, adopted brain imaging technologies, and corresponding brain activation patterns. Also, this study provides practical recommendations for future experiment protocols. Results: The findings indicate that EEG is the preferred neuroimaging sensor for detecting brain activity patterns, compared to fMRI, fNIRS, and PET. Walking is the most studied human locomotion task, likely due to its fundamental nature and status as a reference task. In contrast, running has received little attention in research. Additionally, cycling on an ergometer at a speed of 60 rpm using fNIRS has provided some research basis. Dual-task walking tasks are typically used to observe changes in cognitive function. Moreover, research on locomotion has primarily focused on healthy individuals, as this is the scenario most closely resembling free-living activity in real-world environments. Discussion: Finally, the paper outlines the standards and recommendations for setting up future experiment protocols based on the review findings. It discusses the impact of neurological and musculoskeletal factors, as well as the cognitive and locomotive demands, on the experiment design. It also considers the limitations imposed by the sensing techniques used, including the acceptable level of motion artifacts in brain-body imaging experiments and the effects of spatial and temporal resolutions on brain sensor performance. Additionally, various experiment protocol constraints that need to be addressed and analyzed are explained.

Inertia-Constrained Reinforcement Learning to Enhance Human Motor Control Modeling.

Locomotor impairment is a highly prevalent and significant source of disability and significantly impacts the quality of life of a large portion of the population. Despite decades of research on human locomotion, challenges remain in simulating human movement to study the features of musculoskeletal drivers and clinical conditions. Most recent efforts to utilize reinforcement learning (RL) techniques are promising in the simulation of human locomotion and reveal musculoskeletal drives. However, these simulations often fail to mimic natural human locomotion because most reinforcement strategies have yet to consider any reference data regarding human movement. To address these challenges, in this study, we designed a reward function based on the trajectory optimization rewards (TOR) and bio-inspired rewards, which includes the rewards obtained from reference motion data captured by a single Inertial Moment Unit (IMU) sensor. The sensor was equipped on the participants' pelvis to capture reference motion data. We also adapted the reward function by leveraging previous research on walking simulations for TOR. The experimental results showed that the simulated agents with the modified reward function performed better in mimicking the collected IMU data from participants, which means that the simulated human locomotion was more realistic. As a bio-inspired defined cost, IMU data enhanced the agent's capacity to converge during the training process. As a result, the models' convergence was faster than those developed without reference motion data. Consequently, human locomotion can be simulated more quickly and in a broader range of environments, with a better simulation performance.

Development of a novel precision instrument for high-resolution simultaneous normal and shear force measurements between small planar samples.

In the design and development of end effector pads for silicon wafer handling robots, it is imperative that the static friction/adhesion force properties of the pads with respect to a variety of planar surfaces be characterized. In this work, the overall design, calibration, and data acquisition procedure of an instrument developed for performing these measurements on small (<10 mm × 10 mm) planar samples is presented. This device was used to perform adhesion/maximum shear force measurements on polydimethylsiloxane, a silicon wafer, and custom carbon nanotubes forest surfaces. The device was successfully able to measure an effective, mean profile adhesion force of 715 µN between a silicon wafer and a polydimethylsiloxane (2.768 × 10-6 m2) sample. In addition, a nonlinear maximum shear over normal force relationship was also measured between custom carbon nanotubes forest and the silicon wafer surfaces. The maximum shear over a normal force coefficient was found to decrease with increasing initial normal force. Currently, there are numerous devices for measuring normal/shear forces at the nano/micro- and macroscales; however, this device allows for the consistent measurement of these same types of forces on components with surface dimensions ranging from 0.1 mm to 10 mm.

Highly Anisotropic Adhesive Film Made from Upside-Down, Flat, and Uniform Vertically Aligned CNTs.

We have created a multifunctional dry adhesive film with transferred vertically aligned carbon nanotubes (VA-CNTs). This unique VA-CNT film was fabricated by a multistep transfer process, converting the flat and uniform bottom of VA-CNTs grown on atomically flat silicon wafer substrates into the top surface of an adhesive layer. Unlike as-grown VA-CNTs, which have a nonuniform surface, randomly entangled CNT arrays, and a weak interface between the CNTs and substrates, this transferred VA-CNT film shows an extremely high coefficient of static friction (COF) of up to 60 and a shear adhesion force 30 times higher (12 N/cm2) than that of the as-grown VA-CNTs under a very small preloading of 0.2 N/cm2. Moreover, a near-zero normal adhesion force was observed with 20 mN/cm2 preloading and a maximum 100-µm displacement in a piezo scanner, demonstrating ideal properties for an artificial gecko foot. Using this unique structural feature and anisotropic adhesion properties, we also demonstrate effective removal and assembly of nanoparticles into organized micrometer-scale circular and line patterns by a single brushing of this flat and uniform VA-CNT film.

A Microfluidic Platform Containing Sidewall Microgrooves for Cell Positioning and Trapping.

Microfluidic channels enable the control of cell positioning and the capturing of cells for high-throughput screening and other cellular applications. In this paper, a simple microfluidic platform is proposed for capturing small volumes of cells using sidewall microgrooves. The cell docking patterns in the channels containing sidewall microgroove are also studied. Both numerical and experimental investigations are performed within channels containing sidewall microgrooves of three different widths (i.e., 50, 100 and 200 µm). It is observed that channels containing sidewall microgrooves play an important role in regulating cell positioning and patterning. The obtained results revealed that 10 to 14 cells were positioned inside the sidewall channels of 200 µm width, two to five cells were positioned within the channels of 100 µm width, and one to two individual cells were docked within the sidewall channel of 50 µm width. Particle modelling shows the prediction of cell positioning within sidewall microgrooves. The positions of cells docked within microgroove-containing channels were also quantified. Furthermore, the shear stress variation and cell positioning in the sidewall microgrooves were correlated. Therefore, these sidewall microgroove-containing channels could be potentially useful for regulating cell positioning and patterning on two-dimensional surfaces, three-dimensional microenvironments and high-throughput screening. Cell patterning and positioning are of great importance in many biological applications, such as drug screening and cell-based biosensing.

A response to the commentary on the paper "A comprehensive modeling and vibration analysis of AFM microcantilevers subjected to nonlinear tip-sample interaction forces" by Sohrab Eslami and Nader Jalili".

Recently we received a commentary by Passian et al. [1] on our previously published paper [2] in the Ultramicroscopy. Although the commentators, who are also our co-authors in an earlier papers in similar line (see [3,5]), claimed to discuss and better explain the concept of "virtual resonance" in [3], they went on to conclude that the experimental results and findings in [3] have been misrepresented in [2]. We have theoretically considered a general semi-empirical nonlinear interaction force consisting of different terms with which the dynamic response of the system could be very dependent. Here, we highlight these types of dependencies on the amplitude at the difference frequency. Due to the nature of this communication, we would like to limit our response to matters of fact to the authors of the commentary.

A self-sensing piezoelectric microcantilever biosensor for detection of ultrasmall adsorbed masses: theory and experiments.

Detection of ultrasmall masses such as proteins and pathogens has been made possible as a result of advancements in nanotechnology. Development of label-free and highly sensitive biosensors has enabled the transduction of molecular recognition into detectable physical quantities. Microcantilever (MC)-based systems have played a widespread role in developing such biosensors. One of the most important drawbacks of all of the available biosensors is that they all come at a very high cost. Moreover, there are certain limitations in the measurement equipments attached to the biosensors which are mostly optical measurement systems. A unique self-sensing detection technique is proposed in this paper in order to address most of the limitations of the current measurement systems. A self-sensing bridge is used to excite piezoelectric MC-based sensor functioning in dynamic mode, which simultaneously measures the system's response through the self-induced voltage generated in the piezoelectric material. As a result, the need for bulky, expensive read-out equipment is eliminated. A comprehensive mathematical model is presented for the proposed self-sensing detection platform using distributed-parameters system modeling. An adaptation strategy is then implemented in the second part in order to compensate for the time-variation of piezoelectric properties which dynamically improves the behavior of the system. Finally, results are reported from an extensive experimental investigation carried out to prove the capability of the proposed platform. Experimental results verified the proposed mathematical modeling presented in the first part of the study with accuracy of 97.48%. Implementing the adaptation strategy increased the accuracy to 99.82%. These results proved the measurement capability of the proposed self-sensing strategy. It enables development of a cost-effective, sensitive and miniaturized mass sensing platform.

Integrated automated nanomanipulation and real-time cellular surface imaging for mechanical properties characterization.

Surface microscopy of individual biological cells is essential for determining the patterns of cell migration to study the tumor formation or metastasis. This paper presents a correlated and effective theoretical and experimental technique to automatically address the biophysical and mechanical properties and acquire live images of biological cells which are of interest in studying cancer. In the theoretical part, a distributed-parameters model as the comprehensive representation of the microcantilever is presented along with a model of the contact force as a function of the indentation depth and mechanical properties of the biological sample. Analysis of the transfer function of the whole system in the frequency domain is carried out to characterize the stiffness and damping coefficients of the sample. In the experimental section, unlike the conventional atomic force microscope techniques basically using the laser for determining the deflection of microcantilever's tip, a piezoresistive microcantilever serving as a force sensor is implemented to produce the appropriate voltage and measure the deflection of the microcantilever. A micromanipulator robotic system is integrated with the MATLAB(®) and programmed in such a way to automatically control the microcantilever mounted on the tip of the micromanipulator to achieve the topography of biological samples including the human corneal cells. For this purpose, the human primary corneal fibroblasts are extracted and adhered on a sterilized culture dish and prepared to attain their topographical image. The proposed methodology herein allows an approach to obtain 2D quality images of cells being comparatively cost effective and extendable to obtain 3D images of individual cells. The characterized mechanical properties of the human corneal cell are furthermore established by comparing and validating the phase shift of the theoretical and experimental results of the frequency response.

A comprehensive modeling and vibration analysis of AFM microcantilevers subjected to nonlinear tip-sample interaction forces.

Precise and accurate representation of an Atomic Force Microscopy (AFM) system is essential in studying the effects of boundary interaction forces present between the probe's tip and the sample. In this paper, a comprehensive analytical model for the AFM system utilizing a distributed-parameters based approach is proposed. More specifically, we consider two important attributes of these systems; namely the rotary inertia and shear deformation when compared with the Euler-Bernoulli beam theory. Moreover, a comprehensive nonlinear interaction force is assumed between probe's and sample in order to reveal the response of the system more realistically. This nanoscale interaction force is based on a general form consisting of both attractive and repulsive components as well as a function of the tip-sample distance and the microcantilever's base and sample oscillations. Mechanical properties of the sample could interact with the nanomechanical coupling field between the probe' tip and sample and be implemented in studying the composition information of the sample and the ultra-small features inside it. Therefore, by modulating the dynamics of the AFM system such as the driving amplitude of the microcantilever the procedure for the subsurface imaging is described. The presented approach here could be implemented for designing the AFM probes by examining the tip-sample interaction forces dominant by the van der Waals forces. Several numerical case studies are presented and the force-distance diagram reveals that the proposed nonlinear nanomechanical force along with the distributed-parameters model for the microcantilever is able to fulfill the mechanics of the Lennard-Jones potential.

Robust adaptive backstepping control of uncertain Lorenz system.

In this paper, a novel robust adaptive control method is proposed for controlling the Lorenz chaotic attractor. A new backstepping controller for the Lorenz system based on the Lyapunov stability theorem is proposed to overcome the singularity problem that appeared in using the typical backstepping control method. By exploiting the property of the system, the resulting controller is shown to be singularity free and the closed loop system is globally stable. Due to unavailability of system states measurement in practice, the controller is selected such that only one system state is needed. To overcome the problem of parameter uncertainty, an additional term to Lyapunov function is added and an identification scheme is adopted to have a negative definite Lyapunov function derivative. The simulation results demonstrate the effectiveness of the proposed controllers and approaches.

Development, analysis and control of a high-speed laser-free atomic force microscope.

This paper presents the development and control of a laser-free atomic force microscopy (AFM) system for high-speed imaging of micro- and nanostructured materials. The setup uses a self-sensing piezoresistive microcantilever with nanometer accuracy to abolish the need for a bulky and expensive laser measurement system. A basic model for the interaction dynamics of AFM tip and sample in the high-speed open-loop imaging mode is proposed, accounting for their possible separation. The effects of microcantilever and sample stiffness and damping coefficients on the accuracy of imaging are studied through a set of frequency-domain simulations. To improve the speed of operation, a Lyapunov-based robust adaptive control law is used for the AFM XY scanning stage. It is shown that the proposed controller overcomes the frequency limits of the PID (Proportional-Integral-Derivative) controllers typically used in AFM. Finally, the paper presents a set of experiments on a standard calibration sample with 200 nm stepped topography, indicating accurate imaging up to the scanning frequency of 30 Hz.