A wind power short-term forecasting method based on discrete wavelet transform and long memory networks (DWT_LSTM) is proposed. The LSTM network designed to effectively exhibit the dynamic behavior of time series. introduced decompose non-stationary series into several components which have more stationarity are easier predict. Each component dug by an independent LSTM. results obtained synthesizing prediction values all components. accuracy has been improved proposed method, validated MAE...

State-of-the-art exoskeletons are typically limited by the low control bandwidth and small-range stiffness of actuators, which based on high gear ratios elastic components (e.g., series actuators). Furthermore, most discrete gait phase detection and/or control, resulting in discontinuous torque profiles. To fill these two gaps, we developed a portable, lightweight knee exoskeleton using quasi-direct-drive (QDD) actuation that provides 14 N·m (36.8% biological joint moment for overground...

Predicting human motion plays a crucial role in ensuring safe and effective human-robot close collaboration intelligent remanufacturing systems of the future. Existing works can be categorized into two groups: those focusing on accuracy, predicting single future motion, generating diverse predictions based observations. The former group fails to address uncertainty multi-modal nature while latter often produces sequences that deviate too far from ground truth or become unrealistic within...

This article presents a disassembly task planning algorithm considering human–robot collaboration (HRC) and human behavior prediction (HBP). Unlike assembly procedures, the of end-of-life (EOL) products has been labor-intensive process with uncertainties difficult to cope with. Meanwhile, it is usually challenging obtain an optimal sequence efficiently without excessive computational cost. Also, conventional human-centered planning, in which robot halt frequently due unsafe interruptions by...

Quadrotor unmanned aerial vehicles (UAVs) have been developed and applied into several types of workplaces, such as warehouses, which usually involve human workers. The co-existence UAVs brings new challenges to UAVs: potential failure may cause risk danger surrounding human. Effective efficient detection provide early warning the workers reduce beings much possible. One most common reasons that UAV's flight is physical damage propellers. This paper presents a method detect propellers only...

As unbalance state of charge (SOC) storage units usually leads to the decrease lifetime, SOC balancing control is essential. In this article, a decentralized method proposed balance cascaded-type energy systems. Since does not rely on any communication, it possesses higher reliability. well known, slowly changing variable compared other variables such as voltage and current. Thus, studied system has obvious two-time-scale characteristic. addition, stability analysis based singular...

While the disassembly of high-precision electronic devices is a predominantly labor-intensive process, collaborative robots provide promising solution through human–robot collaboration. To ensure efficient yet safe collaboration, this article presents new way to generate task-constrained and collision-free motion for robot operating in dynamic environment involving human movement, which traditionally challenging due high degree freedom corobot uncertainty nature motion. We first establish...

Ensuring the safety of human workers in a collaborative environment with robots is utmost importance. Although accurate pose prediction models can help prevent collisions between and robots, they are still susceptible to critical errors. In this study, we propose novel approach called deep ensembles temporal graph neural networks (DE-TGN) that not only accurately forecast motion but also provide measure uncertainty. By leveraging employing stochastic Monte-Carlo dropout sampling, construct...

This research introduces a virtual reality (VR) training system for improving human–robot collaboration (HRC) in industrial disassembly tasks, particularly e-waste recycling. Conventional approaches frequently fail to provide sufficient adaptability, immediate feedback, or scalable solutions complex workflows. The implementation leverages Quest Pro’s body-tracking capabilities enable ergonomic, immersive interactions with planned eye-tracking integration improved interactivity and accuracy....

Vibration suppression is of fundamental importance to the performance industrial robot manipulators. Cost constraints, however, limit design options servo and sensing systems. The resulting low drive-train stiffness lack direct load-side measurement make it difficult reduce vibration robot's end-effector hinder application manipulators many demanding applications. This paper proposes a few ideas iterative learning control (ILC) for Compared state-of-the-art techniques such as dual-stage ILC...

The LLC resonant converter can be used as a dc-dc transformer (DCX) to provide isolation and fixed voltage transfer ratio. For applications requiring bidirectional power transmission, this paper proposes PWM modulation control strategy achieve naturally flow for LLC-DCX, even in the case when parameters vary. Without loss of generality, proposed ZVS one side switches ZCS other switches. Meanwhile, additional benefits are that logic exchanging primary full bridge secondary is avoided...

This article presents a generalized disturbance observer (DOB) design framework that is applicable to both multi-input multi-output (MIMO) and/or nonminimum phase systems. The removes conventional DOB's structure constraint, which allows minimizing the H-infinity norm of dynamics from its estimation error over larger feasible set. procedure does not require explicit plant inverse, usually challenging obtain for MIMO or Furthermore, DOB augmented by learning scheme, motivated iterative...

Constant power loads (CPLs) may cause instability of direct-current (dc) distribution networks. This article studies the existence and stability equilibria dc-distribution networks where sources are buck converters under droop control, cables resistances inductances, ZIP (constant impedance, constant current, power) loads. Particularly, a sufficient necessary condition that guarantees is determined. Then, conditions to guarantee system has at least one stable equilibrium First, for...

This paper presents the controller design for path following of a spherical mobile robot, BHQ-1. Firstly, desired velocity reference is deduced from kinematic model, which cannot be transformed into classic chained form. Secondly, necessary torque obtained based on dynamic model. As to kinematics, one-dimensional function selected measure two-directional tracking error, and rolling forward reasonably assumed constant; therefore multiple-input multiple-output (MIMO) system single-input...

Road adhesion coefficient is an important parameter in vehicle active safety control system. Many researchers estimate road by total tire self-aligning torque (SAT, also called front-axle aligning torque), which obtains the average of front wheels, thus leading large estimation error. In this paper, a novel based on single SAT, obtained distribution, brought forward. Due to use proposed method available steering only condition. The main idea that estimated SAT instead SAT. closer real state,...

The evaluation of driver's steering comfort, which is mainly concerned with the haptic driver-vehicle interaction, important for optimization advanced driver assistance systems. current approaches to investigating comfort are based on subjective evaluation, time-consuming, expensive, and easily influenced by individual variations. This paper makes some tentative investigation objective electromyogram (EMG) movement trajectory upper limbs during maneuvers. First, a experiment 21 subjects...

The safety of human workers has been the main concern in human-robot close collaboration. Along with rapidly developed artificial intelligence techniques, deep learning models using two-dimensional images have become feasible solutions for motion detection. These serve as "sensors" closed-loop system that involve humans and robots. Most existing methods detect do not consider uncertainty from model itself. mappings established by should be taken blindly, thus a natural part this type sensor....

Human motion prediction is the foundation stone of human–robot collaboration in intelligent manufacturing. The nonlinear and stochastic nature human has made it challenging to predict accurately. Many recent deep-learning-based approaches, e.g., convolutional neural networks or recurrent (RNNs), have been applied address this challenge. On other hand, existing works tend ignore importance dynamics prediction, especially effect muscle force on motion. This article proposes a novel dynamic...

Most industrial robots are actuated using geared motors with no direct load side measurement. The flexibility introduced by the gear reducer causes transmission errors and vibrations, which limits adoption of robot manipulators in many demanding applications. This paper presents a lean efficient scheme iterative learning control (ILC) to compensate for joint manipulators. A two-degree-of-freedom ILC method is introduced. Compared dual-stage that has been previously proposed servo...

A set of in-vessel saddle coils called dynamic resonant magnetic perturbation (DRMP) for generating rotating perturbations has recently been constructed on the J-TEXT tokamak. The phenomenon tearing mode locking to DRMP and together with field observed. There is an apparent decrease island width during unlocking procedure. Similar results are obtained in numerical simulation.