Over the past five years, the fast development and use of generative artificial intelligence (GenAI) and large language models (LLMs) has ushered in a new era of study, teaching, and learning in many domains. The role that GenAIs can play in engineering research is addressed. The related previous works report on the potential of GenAIs in the literature review process. However, such potential is not demonstrated by case studies and practical examples. The previous works also do not address how GenAIs can assist with all the steps traditionally taken to design research. This study examines the effectiveness of collaboration with GenAIs at various stages of research design. It explores whether collaboration with GenAIs can result in more focused and comprehensive outcomes. A generalised approach for collaboration with AI tools in research design is proposed. A case study to develop a research design on the concept of “shared machine-human driving” is deployed to show the validity of the articulated concepts. The case study demonstrates both the pros and cons of collaboration with GenAIs. The results generated at each stage are rigorously validated and thoroughly examined to ensure they remain free from inaccuracies or hallucinations and align with the original research objectives. When necessary, the results are manually adjusted and refined to uphold their integrity and accuracy. The findings produced by the various GenAI models utilized in this study highlight the key attributes of generative artificial intelligence, namely speed, efficiency, and scope. However, they also underscore the critical importance of researcher oversight, as unexamined inferences and interpretations can render the results irrelevant or meaningless.

Road intersections will soon be congested with countless connected autonomous vehicles (CAVs) to meet a variety of transportation needs. In such an environment, CAVs will need to work together and coordinate with one another. Smart infrastructure should also be in place to enable CAVs to efficiently utilize shared intersection resources and carry out their mobility duties. The interaction among CAVs and between CAVs and infrastructure for efficient intersection management is facilitated by a key technology of intelligent transportation system (ITS), known as cooperative vehicle-intersection system (CVIS). Towards developing a better insight into this fast-developing technology, this study presents a thorough review of the state of the art in CVIS research in the context of three hierarchal layers of shared perception, intersection control, and vehicle control. To describe the workflow of cooperative perception systems, a systematic architecture of infrastructure-based perception supported by edge computing and multi-node multi-sensor fusion techniques is explored. Different shared perception methodologies, aiming at the perceptual fusion of CAVs and infrastructure to achieve comprehensive environmental perception are critically analyzed to identify the landscape of cooperative detection and tracking. With an emphasis on essential interaction between CAVs and infrastructure, various cooperative control strategies are then investigated for the co-design of crossing scheduling and motion control of multiple CAVs at intersections. The applications of CVIS in cooperative driving automation in terms of efficiency, safety, and sustainability are qualitatively analyzed. This paper concludes by identifying gaps and challenges in vehicle-intersection cooperation along with recommendations on future research directions.

Early detection of driver mental fatigue is one of the active areas of research in smart and intelligent vehicles. There are various methods, based on measuring the physiological characteristics of the driver utilising sensors and computer vision, proposed in the literature. In general, driver behaviour is unpredictable that can suddenly change the nature of driving and dynamics under mental fatigue. This results in sudden variations in driver body posture and head movement, with consequent inattentive behaviour that can end in fatal accidents and crashes. In the process of delineating the different driving patterns of driver states while active or influenced by mental fatigue, this paper contributes to advancing direct measurement approaches. In the novel approach proposed in this paper, driver mental fatigue and drowsiness are measured by monitoring driver’s head posture motions using XSENS motion capture system. The experiments were conducted on 15 healthy subjects on a MATHWORKS driver-in-loop (DIL) simulator, interfaced with Unreal Engine 4 studio. A new modified bidirectional long short-term memory deep neural network, based on a rectified linear unit layer, was designed, trained and tested on 3D time-series head angular acceleration data for sequence-to-sequence classification. The results showed that the proposed classifier outperformed state-of-art approaches and conventional machine learning tools, and successfully recognised driver’s active, fatigue and transition states, with overall training accuracy of 99.2%, sensitivity of 97.54%, precision and F1 scores of 97.38% and 97.46%, respectively. The limitations of the current work and directions for future work are also explored.

Driver mental fatigue is a major influential factor that results in inattentive driver condition ultimately to fatal crashes. In this paper, the driver mental fatigue patterns based on the driver’s body postural behaviour are identified using a novel adaptive pattern recognition technique. The experiments were conducted on 20 healthy subjects in a MATHWORKS simulated driving environment. The posture of the driver was measured using the XSENS motion capture system. To monitor the actions performed under the influence of mental fatigue, variations in the acceleration of the head, neck, and sternum were extracted and deployed in an unsupervised manner. A fully adaptive version of the symbolic aggregate approximation algorithm based on unsupervised clustering was developed that identifies the time-series patterns of driver fatigue posture. The time-variant fatigue patterns were dynamically segmented and symbolized according to the discrepancy in the postural behaviour. The experimental results indicate that the proposed algorithm successfully detects the time-variant fatigue patterns of multivariate dataset compared to the original symbolic pattern recognition tool. The limitations of the current approach and future work in improving driver safety are explored.

Driver's foot behaviour is unpredictable and can suddenly change the nature of driving and dynamics under the influence of different factors that stimulates the driving style. Such effects result in sudden variations in foot dynamics and trajectory between accelerator and brake pedals inducing vagueness in smart active control system. This paper is an extension to the intrusive approach where driver's foot trajectory and shifting between pedals are monitored using XSENS motion capture system. The main objective is to predict the foot patterns associated with acceleration and braking. The experiments were conducted on 10 young subjects on MATHWORKS driver-in-loop (DIL) simulator, interfaced with Unreal Engine 4 studio. A new modified bidirectional long short-term memory (Bi-LSTM) deep neural network based on a rectified linear unit layer was designed, trained, tested and compared with traditional machine learning algorithms on 3D time-series foot orientation data for the sequence-to-sequence classification. The results show that the proposed classifier performs well and successfully recognizes the driver's foot behaviour with overall accuracy of 99.8%. Such identified patterns will help in determining the foot posture and the degree of intention in pressing the particular pedal. Moreover, the patterns will be useful for early intervention by smart systems to cope with the longitudinal mistakes made during driving. The limitations of the current work and directions for future work are explored.

In a Steer-by-Wire (SbW) system, the mechanical linkages which connect the steering wheel to front wheels are replaced by a digitally controlled steering system. In spite of improved vehicle safety due to better steering capability, the SbW system actuator failure may lead to unwanted steering performance or even instability. A fault tolerant model predictive control (MPC) with fault compensation for SbW systems based on delta operator for actuator faults is proposed. It deploys an observer to estimate both the fault information and the faulty SbW system states. At each sampling time, the MPC immediately compensates for the fault. The gains of the observer and the fault tolerant MPC controller are obtained by solving a linear matrix inequality derived from the Lyapunov theory. The simulation results illustrate that the proposed fault tolerant control strategy can counter the various types of actuator faults and maintain superior steering performance to shift operator-based MPC controller.

The Steer-by-Wire (SbW) system is an electronically controlled steering system that is able to improve steering capability without mechanical links between the steering wheel and the front wheels. However, failure of the SbW system actuator may lead to steering performance degradation and result in instability. In this paper, a fault tolerant sliding mode predictive control (SMPC) strategy for an SbW system is proposed. The sliding mode control is applied to improve the robustness of the model predictive control (MPC) in the presence of modeling uncertainties and disturbances, while the MPC is applied to enhance the fault tolerant capability of the steering control processes. The chaos particle swarm optimization (CPSO) algorithm is introduced to optimize the MPC and a two-stage Kalman filter is introduced to simultaneously provide fault information and state estimation. The performance of the proposed approach is validated through computer simulation. The results demonstrate that the proposed SMPC-CPSO controller is more robust and provides better tracking performance in the presence of model uncertainties, disturbance, and actuator faults than SCMP-PSOs (heterogeneous comprehensive learning particle swarm optimization, evolutionary particle swarm optimizer, etc), SMPC-differential evolution, MPC, SMPC, and MPC-PSO.

Detection of motor intention from brain signals combined with robot assistive technologies has potential to be used as an effective rehabilitation process for post-stroke patients. The work conducted on the deployment of AMADEO hand rehabilitation robotic device and Electroencephalogram based Brain Computer Interference (EEG-BCI) system to explore the technical feasibility of the approach in hand motor recovery of post-stroke patients is presented. Two different protocols consisting of simple visual cues and a 2D interactive game are presented to healthy subjects when performing hand movement. The motor intent signals produced during each protocol are detected using Support Vector Machine (SVM) algorithm. Moreover, the signals produced by different single electrodes are analyzed to identify the electrode making the highest contribution to the intent signal and the performance of SVM with respect to each protocol. Overall, an average True Positive Rate (TPR) of 71.72% and True Negative Rate (TNR) of 63.33% for visual cue protocol and an average TPR of 88.56% and TNR of 70.81% for game protocol are obtained.

Assessing the frailty of older people quantitatively is critical to prevent potential accidents and to ensure their well-being. The older people with high frailty score are at the risk of fall, which increases the rate of hospitalization and reducing the number of independent activities carried out. The conventional clinical tools used for frailty assessment are subjective, qualitative and are prone to human error. The balance assessment, activity of daily living (ADL) and gait analysis are practiced as clinical and quantitative tools for risk of fall and frailty assessments. An objective approach to classify the frailty levels using ADL is proposed. The "pick up an object from floor" as an ADL is deployed to differentiate the signal patterns obtained through inertial measurement unit (IMU) for frail and non-frail subjects. The data from single inertial unit mounted on pelvis is analyzed. The experimental work is carried out on three groups of healthy/control, frail and non-frail subjects. The various signal attributes are used to classify the frailty quantitatively using IMU data and machine learning methods. The results demonstrate that frail subjects have clear irregularities in their signal trajectories. Using the proposed algorithm two classes of frailty (non-frail and frail) are identified objectively. The study demonstrates the potential of deploying IMU for advanced classification of frailty levels in older people.

In this paper, a novel neural network (NN)-based four-channel wave-based time domain passivity approach (TDPA) is proposed for a teleoperation system with time-varying delays. The designed wave-based TDPA aims to robustly guarantee the channels passivity and provide higher transparency than the previous power-based TDPA. The applied NN is used to estimate and eliminate the system's dynamic uncertainties. The system stability with linearity assumption on human and environment has been analyzed using Lyapunov method. The proposed algorithm is validated through experimental work based on a 3-DOF bilateral teleoperation platform in the presence of different time delays.

This paper presents an approach for learning robust models of humanoid robot trajectories from demonstration. In this formulation, a model of the joint space trajectory is represented as a sequence of motion primitives where a nonlinear dynamical system is learned by constructing a hidden Markov model (HMM) predicting the probability of residing in each motion primitive. With a coordinated mixture of factor analyzers as the emission probability density of the HMM, we are able to synthesize motion from a dynamic system acting along a manifold shared by both demonstrator and robot. This provides significant advantages in model complexity for kinematically redundant robots and can reduce the number of corresponding observations required for further learning. A stability analysis shows that the system is robust to deviations from the expected trajectory as well as transitional motion between manifolds. This approach is demonstrated experimentally by recording human motion with inertial sensors, learning a motion primitive model and correspondence map between the human and robot, and synthesizing motion from the manifold to control a 19 degree-of-freedom humanoid robot.

A survey of the adaptive controllers deployed to address major inherent control issues in robotic teleoperation systems is carried out. The study in particular explores the application of adaptive controllers in dealing with master and slave model uncertainties, operator and environment force model uncertainties, unknown external disturbances, and communication delay. The reviewed literature is structured according to the objectives envisaged for the adaptive controllers. Meanwhile, some adaptive methods deployed in human-robot interaction, where robots collaborate with people and actively support them, and local robot control, where robot manipulators are controlled at the same location as the operator, are also considered in the review as they can be used in teleoperation with some minor adjustment. A comparison of the strengths, deficiencies, and requirement of methods in each category is carried out. The study indicates that the majority of the proposed methods either require additional hardware such as sensors, or assume an accurate model of the system under study. The possible future research directions are outlined based on the gaps identified in the survey.

This paper describes the implementation of a haptic system for simulating the force-feedback of a centuries-old musical instrument, the carillon. A carillonneur's performance depends on a developing familiarity with force-feedback associated with a particular instrument. Force-feedback may vary not only across the range of the instrument but from one instrument to another. However, carillonneurs have restricted access to their instrument for rehearsal, unlike musicians who specialise in other instruments. At best, rehearsal times on the actual instrument are shared with other carillonneurs; at worst, a carilloneur rehearses mainly on a practice instrument that does not prepare them for the force-feedback and sonic characteristics of the actual instrument. A haptic device that simulates the force-feedback of a carillon must meet exacting requirements associated with musical performance and musical skill acquisition. Our haptic device uses a voice-coil linear actuator in a position-control loop where the force applied by a carillonneur is measured by analysing the noisy current produced from an analog servo controller reacting to position error. A parametised model of the servo is determined using system identification methods and forms the basis of a Kalman filter for the servo current signal, closing an admittance display loop around carillonneur-applied forces and the mechanical action of the instrument simulated in a virtual environment. This generalised method is particularly applicable to scenarios common in creative applications where hardware is `handed-down' from other projects.

A new approach is explored to transfer human manipulation skills to a robotics system. A skill acquisition algorithm utilizes the position and contact force/torque data generated in the virtual environment combined with a priori knowledge about the task to generate the skills required to perform such a task. Such skills are translated into actual robotic trajectories for implementation in real time. The peg-in-hole insertion problem is used as a case study. The results are reported.

The work carried out to explore the feasibility of reconstructing human constrained motion manipulation skills is reported. This is achieved by tracing and learning the manipulation performed by a human operator in a haptic rendered virtual environment The peg-in-hole insertion problem is used as a case study. In the developed system, position and contact force and torque as well as orientation data generated in the haptic rendered virtual environment combined with a priori knowledge about the task are used to identify and learn the skills in the newly demonstrated task. The data obtained from the virtual environment is classified into different cluster sets using a competitive fuzzy clustering algorithm called Competitive Agglomeration (CA). The CA algorithm starts with an over specified number of clusters which compete for feature points in the training procedure. Clusters with small cardinalities lose the competition and gradually vanish. The optimal number of clusters that win the competition is eventually determined. The clusters in the optimum cluster set are tuned using Locally Weighted Regression (LWR) to produce prediction models for robot trajectory performing the physical assembly based on the force/position information received from the rig. A background on the work and its significance is provided. The approach developed is explained and the results obtained so far are presented.

The paper carried out to explore the feasibility of reconstructing human constrained motion manipulation skills is reported. This is achieved by tracing and learning the manipulation performed by a human operator in a haptic rendered virtual environment. The peg-in-hole insertion problem is used as a case study. In the developed system, force and position variables generated in the haptic rendered virtual environment combined with a priori knowledge about the task are used to identify and learn the skills in the newly demonstrated task. The data obtained from the virtual environment is classified into different cluster sets using fuzzy Gustafson-Kessel model (FGK). Principal component analysis (PCA) is applied to each cluster to reduce the dimension of the data. The clusters in the optimum cluster set are tuned using locally weighted regression (LWR) to produce prediction models for robot trajectory performing the physical assembly based on the force/position information received from the rig

Focuses on the benchmark control problems for seismically excited nonlinear buildings defined by Ohtori et al. (2000). This benchmark study focuses on three typical steel structures, 3-, 9- and 20-storey buildings designed for the SAC project for Los Angeles in the California region. The first stage of applying the fuzzy controller to this benchmark study for the 3-storey building is reported. The main advantage of the fuzzy controller is its inherent robustness and ability to handle the non-linear behaviour of the structure. This benchmark study is based on a number of evaluation criteria and control constraints and these limitations are considered in the design of the fuzzy controller. The performance of the controller is validated through the computer simulation on MATLAB. The results of the simulation show a good performance of the fuzzy controller to reduce the response of the building under different earthquake excitations.

In compliant motion, the task to be performed is usually not well structured and uncertainty exists. The operational environment is either partially known or unpredictable. In applications such as the manipulation of flexible materials, the characteristics of the plant changes during operation. Conventional control methods, therefore, do not provide an appropriate solution for such problems. Intelligent control (IC), in which neural networks and fuzzy control are key components, is employed to produce a self-learning compliant motion. The neuro-fuzzy model of the compliant motion is obtained through Adaptive Spline Modelling of Observation Data (ASMOD) algorithm. This is used as the initial model of the process. An adaptive indirect fuzzy controller is then designed to control and adapt the system parameters on-line. The results are compared with previous work which employed static fuzzy control.

An approach is described for design and development of robotic system controllers. The technique, referred to as multiprocessing control, can provide more expansibility, flexibility, and portability in a robotic system than conventional multiprocessor methods. The control algorithms for a Mitsubishi robot are implemented as a multiprocessing system on a VME-bus-based computer running the PDOS operating system. The control software is written entirely in a real-time C and consists of concurrent tasks synchronized with events, semaphores, and messages. A proposal for implementing the technique as a hardware-independent software on the Inmos Transputer is given.>