Speed Tracking Research Articles

Enhancing Short Track Speed Skating Performance through Improved DDQN Tactical Decision Model.

This paper studies the tactical decision-making model of short track speed skating based on deep reinforcement learning, so as to improve the competitive performance of corresponding short track speed skaters. Short track speed skating, a traditional discipline in the Winter Olympics since its establishment in 1988, has consistently garnered attention. As artificial intelligence continues to advance, the utilization of deep learning methods to enhance athletes' tactical decision-making capabilities has become increasingly prevalent. Traditional tactical decision techniques often rely on the experience and knowledge of coaches and video analysis methods that require a lot of time and effort. Consequently, this study proposes a scientific simulation environment for short track speed skating, that accurately simulates the physical attributes of the venue, the physiological fitness of the athletes, and the rules of the competition. The Double Deep Q-Network (DDQN) model is enhanced and utilized, with improvements to the reward function and the distinct description of four tactics. This enables agents to learn optimal tactical decisions in various competitive states with a simulation environment. Experimental results demonstrate that this approach effectively enhances the competition performance and physiological fitness allocation of short track speed skaters.

Speed & gear dynamic planning and velocity tracking control considering traffic light information

To optimize fuel economy and improve vehicle longitudinal control performance, this paper proposed a speed & gear planning strategy considering traffic light information and velocity tracking control. First, to guarantee that the vehicle can pass through traffic lights without stop if possible and retain the highest terminal speed, this article discusses a Propelling or Coasting (PoC) serialization scheme. The speed & gear planning are realized by adjusting acceleration while propelling, with the upper and lower speeds set according to vehicle status and traffic light information; Second, to improve the accuracy and anti-perturbation performance of longitudinal control, a compound control method using improved Global Fast Terminal Sliding Mode Control (GFT-SMC) is developed, which enables the reduction of control output while maintaining high tracking performance and smooth control ability. Finally, both the hardware-in-loop (HIL) and real-vehicle experimental results show that the proposed control method effectively improves fuel economy with better speed tracking performance and strong perturbation robustness.

Improved SiamCAR with ranking-based pruning and optimization for efficient UAV tracking

UAV tracking is a burgeoning task with vast application prospects in various fields such as agriculture, navigation, and public safety. However, the computational constraints and limited processing speed of drones hinder the deployment of Siamese tracking algorithms. In order to better apply tracking algorithms to drone devices, this paper proposes a novel Ranking-Based SiamCAR (RB-SiamCAR) tracker. The RB-SiamCAR tracker achieves model compression by utilizing a ranking-based filter pruning method, which sorts the filters in the backbone network based on their importance and prunes the filters with low ranks for efficient feature extraction. Additionally, considering existing Siamese trackers overlook the correlation between positive and negative samples and the coherence between classification and localization, we introduce two ranking-based losses. The classification ranking loss ensures that the ranking of positive samples is higher than that of hard negative samples, allowing the tracker to successfully select foreground samples without being fooled by distractors. The IoU-guided ranking loss aims to align the classification with the intersection over union (IoU) of the corresponding localizations of positive samples, enabling well-localized predictions to be represented by high classification confidence. To further enhance the tracking performance, we employ an effective channel attention module (ECA) that allows the network to automatically learn and focus on the most important channels for capturing more discriminative features. Experimental evaluations on UAV tracking benchmarks demonstrate that the proposed RB-SiamCAR outperforms existing state-of-the-art trackers. It is noteworthy that our RB-SiamCAR achieves an impressive tracking speed of nearly 100 fps. The experimental results validate its effectiveness and efficiency in UAV tracking applications.

Modeling the Early and Late Acceleration Phases of the Sprint Start in Elite Long Track Speed Skaters.

Zukowski, MH, Jordan, MJ, and Herzog, W. Modeling the early and late cceleration phases of the sprint start in elite long track speed skaters. J Strength Cond Res 38(2): 236-244, 2024-This study established the reliability of an exponential function to model the change in velocity during the speed skating sprint start and the validity of associated model parameters in a group of subelite and elite long track speed skaters. Long track speed skaters ( n = 38) performed maximal effort 50-m on-ice accelerations from a standing start while tethered to a horizontal robotic resistance device that sampled position and time data continuously. An exponential function was applied to the raw data to model the change in velocity throughout the acceleration phase and compute the maximal skating speed (MSS), maximal acceleration capacity (MAC), maximum relative net horizontal power ( PMax ), and an acceleration-time constant ( τ ). All constructed models provided a sufficient fit of the raw data ( R -squared > 0.95, mean bias <2%). Intraday reliability of all model parameters ranged from good to excellent (intraclass correlation coefficient >0.8 and coefficient of variation <5%). Strong negative correlations ( r : -0.72 to -0.96) were observed between MSS and PMax and the 10 and 20 m split times measured with the robotic resistance and with 100 split times obtained from 500 m races. Moderate-to-large between-group differences were observed in MSS, MAC, and PMax between the elite vs. subelite speed skaters (Cohen d effect sizes: 1.18-3.53). Our results indicate that monoexponential modeling is a valid and reliable method of monitoring initial acceleration performance in elite level long track speed skaters.

A Fuzzy-Based Proportional–Integral–Derivative with Space-Vector Control and Direct Thrust Control for a Linear Induction Motor

In this study, the analysis and control of a multi-phase linear induction motor loaded with a variable mechanical system are carried out. Mathematical models are established, and simulation results are analyzed for an improved proportional–integral–derivative controller with closed-loop vector control for PLIM. To make the PID controller more responsive to load thrust disturbances, a fuzzy PID load thrust observer was developed. The FPID is similarly based on space-vector modulation DTC technology to regulate the PLIM’s speed, flux, and thrust. The FPID output is used to calculate the reference thrust force, which is compared to the actual thrust value to calculate the second error. To maintain the linear speed of the PLIM at the specified reference values and at different load values, the FPID controller settings are adjusted. Four indicators were used to compare the capabilities of the FPID controller with those of the conventional PID controller in order to evaluate the performance of PLIM in both cases. These indices represent the individual SSE for each operational phase and the total SSE for the entire loading period. According to the simulation results, the FPID works better than a regular PID when used to adjust the operation of DTC-SVM to drive a PLIM to improve the overall system performance. The simulation results using MATLAB Simulink for a PLIM-drive system show that the proposed FPID control provides improved control behavior and operating performance with fast and accurate speed tracking.

A hierarchical reinforcement learning GPC for flexible operation of ultra-supercritical unit considering economy

With the growing proportion of renewable energy sources in power grid, the demands for deep peak shaving and rapid load regulation of ultra-supercritical (USC) unit to provide more flexible and stable power services is further enhanced. Simultaneously, the load tracking speed is no longer the only significant task in the flexible operation of USC unit, the operational economy is more and more important. To this end, a hierarchical reinforcement learning GPC control scheme with a better economy is proposed and apply to USC unit. Firstly, the model mismatch problem in control process is taken into account, and an integral mode is added into the GPC to eliminate the steady state error during the load regulation. Secondly, the hierarchical reinforcement learning GPC scheme integrated the GPC and twin delayed deep deterministic policy gradient algorithms is presented. The complex flexibility demand problem is decomposed into rapid load regulation and economic operation and then respectively solved in the GPC-based upper layer and reinforcement learning agent-based lower layer. Thirdly, under the flexibility demand, a multi-criteria optimization function including the load tracking cost, main steam throttling loss and coal consumption rate of unit is constructed. Then, via the accurate trajectory tracking by GPC and on-line control sequence optimization by reinforcement learning agent, the operational flexibility and economy of USC unit are simultaneously improved within safe boundaries. Finally, to verify the availability of the proposed hierarchical reinforcement learning GPC strategy, large-scale load variations and disturbance rejection ability tests cover the 30%–100 % rated load of a 1000 MW USC unit are carried out, and the load regulation performance is compared to other strategies. Results confirm that the designed hierarchical reinforcement learning GPC control scheme is profit for the flexible and economic operation capability promotion of USC unit.

A study on the application of artificial intelligence in the design of intelligent medical robots

Abstract With the development of artificial intelligence and robotics technology, the combination of artificial intelligence and medical device research and development has been promoted, which is an important product of the development of artificial intelligence. In this paper, the general structure of the intelligent medical robot is designed by combining artificial intelligence technology and robotics-related technology. Then, the binocular vision function of the robot was realized by visually acquiring the image of the target object, 3D reconstruction of the target object, and combining the SIFT image recognition algorithm and target tracking algorithm. Then, a new speech recognition algorithm was constructed to realize the human-robot interaction function with the medical robot based on the deep learning Transforme network after the construction of the human acoustic model. Finally, the designed intelligent medical robot was tested, and its overall performance was evaluated. The results show that the recognition errors of the intelligent medical robot on the features of the items are all within 0.05, the recognition errors on the features of the human body are within 0.2, and the speed of the target tracking is between 6km/h and 16km/h. The average recognition accuracy of the medical robot for voice commands is about 0.9, the recognition time is about 0.7s, the normal working rate of each function is more than 0.99, and the test speed is within 2s.

Design recommendations of target size and tracking speed under circular and square trajectories for smooth pursuit with Euclidean algorithm in eye-control system

In this study, we investigated how the size, speed, and trajectory shape of smooth-pursuit targets for Euclid's algorithm affect the ergonomics and usability of the gaze interface, and conducted a user study with 20 participants, examining the priority of five target sizes in circular and square trajectories, and the recommended speed for each size in the five target speeds. The usability of the SUS and ISO 9241 (acceptability and comfort) of the eye control system had been evaluated. Based on the experiment results, the design recommendations of the priority of the target size and speed of the circular and square trajectories, suggested trajectory shapes. The research is an important reference for the design and development of smooth-pursuit targets based on the Euclidean algorithm.

A Novel Approach for Train Tracking in Virtual Coupling Based on Soft Actor-Critic

The development of virtual coupling technology provides solutions to the challenges faced by urban rail transit systems. Train tracking control is a crucial component in the operation of virtual coupling, which plays a pivotal role in ensuring the safe and efficient movement of trains within the train and along the rail network. In order to ensure the high efficiency and safety of train tracking control in virtual coupling, this paper proposes an optimization algorithm based on Soft Actor-Critic for train tracking control in virtual coupling. Firstly, we construct the train tracking model under the reinforcement learning architecture using the operation states of the train, Proportional Integral Derivative (PID) controller output, and train tracking spacing and speed difference as elements of reinforcement learning. The train tracking control reward function is designed. Then, the Soft Actor-Critic (SAC) algorithm is used to train the virtual coupling train tracking reinforcement learning model. Finally, we took the Deep Deterministic Policy Gradient as the comparison algorithm to verify the superiority of the algorithm proposed in this paper.

Vehicular Platoon Control With Gain Variations via Unreliable VANETs

This paper investigates the adaptive control for vehicular platoon with controller gain variations and packet dropouts. With leader predecessor following topology, vehicle speed tracking and distance keeping are achieved by tracking-error based control. The gain variations are compensated for by introducing an adaptive estimation law. A switching adaptive control law is presented to deal with packet dropouts, in which only distance measurements are used when packet dropouts occur. Combining the switched system approach and adaptive control, a set of stability conditions are obtained to design the adaptive controller for each follower. The results present a unified framework for determining the controller gains and the estimator adaptation law, yielding an adaptive control algorithm that guarantees both internal stability and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\mathcal {L}_{2}$</tex-math></inline-formula> string stability. Numerical examples demonstrate the effectiveness of our method.

Maximum Power Point Tracking Based on Wind Speed Prediction

A control method for MPPT (maximum power point tracking) based on wind speed prediction is proposed to address the issue of slow attainment of optimal speed values by wind turbines in variable wind speed conditions in the wind power system. Initially, the wind speed variation value is accurately estimated using the SSA-BP neural network based on historical wind speed data. Subsequently, the MPP (maximum power point) is calculated considering the optimal blade-tip speed ratio, thereby enhancing the tracking speed of the system. Simulation results demonstrate that this method enables real-time estimation of wind speed variation, improves wind energy utilization rate, and maximizes the utilization of wind energy.

FPGA Implementation of Adaptive PID Control for Quadcopter Position Tracking

AbstractPID control is widely used for designing controllers in production plants and automotive and robotic applications. Although it suits linear systems, with minor modifications, it can also be applied to nonlinear plants. As a simple control approach, it can be easily realized conceptually, but there is a large cost involved in the production of the working controller. A digital PID controller based on a fixed point representation of the operands can be considered for approximate realization. The field programmable gate array (FPGA), with its concurrent architecture, is an ideal way of producing programmable, cost‐, power‐, and speed‐optimized controllers with a compromised controller area compared to application‐specific approaches. In this research, the PID controller for quadcopters is initially tested for performance using MATLAB/Simulink based on continuous/analog domain operations. Then, FPGA‐based architecture mapping is performed using HDL Coder, with fixed‐point‐based operands produced using the fixed‐point conversion toolbox. With satisfactory performance obtained with system generator‐based cosimulation of the controller, Vivado IDE is used to implement the controller on a Zynq Ultrascale+ based FPGA device. The realized cost‐efficient controller is analysed for logical resources, computation power and controller frequency and inferred to be optimal with reference to all these indices. By adapting the thrust, which is the control input for the translational motion of the quadcopter, robustness against coupled motion is ensured, with higher simultaneous tracking speeds.

Backstepping Sliding Mode Control Algorithm for Unmanned Aerial Vehicles Based on Fractional-Order Theory

Aiming at the problems of slow convergence speed and low tracking accuracy in attitude control and position tracking of quadrotor unmanned aerial vehicles (UAVs). This paper combines the fractional-order calculus theory with the backstepping sliding mode control algorithm, using the backstepping control to compensate for the nonlinearity of the system and the fractional-order theory to eliminate the jitter brought about by the sliding mode control, and proposes a new fractional-order backstepping sliding mode control strategy for the trajectory tracking control of the quadrotor UAV. The proposed fractional-order sliding mode surface increases the control flexibility and improves the robustness and anti-interference ability of the system to some extent. The stability analysis of the system is carried out using the Lyapunov stability theory, and the results prove the stability of the proposed controller. Finally, the effectiveness and feasibility of the proposed method are verified by comparing it with the traditional backstepping sliding mode controller. The simulation results show that the fractional-order reverse-step sliding mode control algorithm proposed in this paper is significantly better than other control algorithms in terms of convergence speed and also has a certain degree of superiority in terms of error elimination.

Comparison of data-based and modeled-based analysis of aircraft departure noise using noise monitor network recordings

A comparison of data-based regression analysis and modeled-based analysis of aircraft departure noise variations using airport noise monitor network recordings is presented. Accurate modeling of aircraft noise is useful for the assessment of departure procedures and provides insight into the noise reduction potential of noise abatement flight procedures. In addition, many airports employ monitor networks that continuously record noise data produced by aircraft on departure, which can be used for assessing potential factors leading to variation in departure noise and for comparison to results from modeling methods. In this paper, departure noise variations of Boeing 737-800 aircraft are correlated to departure climb procedure factors using both data-based approaches incorporating recorded noise collected at the Seattle International Airport monitor network, and a method to model departure noise from on altitude, ground track, and ground speed from recorded operational surveillance data. Examples using both approaches indicate altitude and velocity result in the highest correlating factors for noise at the three noise monitor locations studied. Limitations and areas of needed improvement for the different noise modeling methods are highlighted.

Pengenalan Pakaian Adat Aceh Berbasis Augmented Reality Menggunakan Metode Speed Up Robust Featured (SURF)

Aceh Province, especially Langsa City, has a tourist attraction, namely the Langsa City Forest Park House (RTH). One of the most interesting rides in Langsa City Forest Park is Rumoh Aceh. Based on the results of visits and interviews by Rumoh Aceh officers, the large number of visitors from outside Aceh with one officer sometimes made it difficult for the officers to explain the information available about Rumoh Aceh, especially Acehnese traditional clothes, which were only displayed from a printed image, and provided no other information about these traditional clothes, so that many visitors did not know the diversity of designs and motifs of traditional clothes in Aceh. So, a medium was formed that could display Acehnese traditional clothing. The media uses augmented reality technology so that users can add virtual objects to the real environment to make it easier to use. This application uses the Speed Up Robust Featured (SURF) algorithm, which can process marker tracking quickly so that it can obtain better tracking speed times. The shortest distance from the marker to the camera that can show 3D objects is 20 cm, whereas the farthest distance that cannot show 3D objects is 100 cm. The best distance at which a marker can be detected is 20–80 cm. The best average detection time is 0.00049 s, and the average speed obtained is 1261.22 m/s at a distance of 60 cm. The Speed Up Robust Featured (SURF) algorithm can be used in the Augmented Reality-based Aceh Traditional Clothing Recognition application.

Wind turbine output power prediction by a segmented multivariate polynomial-XGBoost model

ABSTRACT Accurate prediction of wind turbine output power is important to ensure safe, stable, and efficient wind power generation. In this study, a hybrid prediction model based on a combination of segmented multivariate polynomial and extreme gradient boosting (XGBoost) under different operating conditions is proposed. First, the wind farm supervisory control and data acquisition (SCADA) data are processed using isolated forest for outliers, in order to improve model prediction accuracy. Then, random forest is used for feature selection, where wind speed, rotor speed, pitch angle, wind direction and temperature are found to be the most relevant features related to the output power. Next, based on the analysis of the wind turbine characteristic curves, its operating state is divided into three phases: constant power (CP), constant speed (CS), and maximum power point tracking (MPPT). Different polynomials are established for each phase, the CP phase is of order 5, the CS phase is of order 7, and the MPPT phase is of order 4. Finally, the XGBoost model is trained for power prediction with polynomial new features for each phase. The segmented multivariate polynomial-XGBoost model has a mean absolute percentage error (MAPE) of 0.04, a root mean square error (RMSE) of 49, and a coefficient of determination R2 of 0.99. Compared to the single XGBoost model and the unsegmented polynomial-XGBoost model, the MAPE decreases by 42% and 33% respectively, while RMSE decreases by 50% and 12% respectively, resulting in high prediction accuracy.

Application of adaptive global terminal sliding mode control based on limited error in high-speed train

High-precision tracking control is very important in high-speed train (HST) operation. This study proposes an adaptive global terminal sliding mode control scheme based on limited error (AEG-SMC) for trajectory tracking of HST. AEG-SMC is firstly proposed for the nonlinear dynamic system to improve the accuracy of tracking error and solve the problem of parameter uncertainty in the model. Then, AEG-SMC uses an adaptive method to approximate the uncertain parameters in the model to obtain real-time parameter estimation, and an error index function is given to realise fast and high-precision tracking of the system by global terminal sliding mode control (GSMC). AEG-SMC is able to design the required error indicators before conducting control, thereby ensuring the high-quality response and stability of the system. Finally, the simulation results show that the speed tracking error and control input of HST with AEG-SMC are also the smallest among the mentioned methods.

Optimal control of DC motor using leader-based Harris Hawks optimization algorithm

Direct Current (DC) motor is considered a very critical component of various industrial drive equipment. This is due to their unique advantages including reasonable cost, speed-torque characteristics, ease of control, etc. While most DC motor drive applications often employ PID controllers to regulate the speed of the machine, selecting the optimal design parameters for the controllers used in these applications often posed a serious challenge. Moreover, as the complexity of the industrial process increases, the need for precise speed and position tracking becomes necessary. This current study proposes a novel Leader-based Harris Hawks Optimization (LHHO) algorithm for the design of Proportional-Integral-Derivative (PID) and Fractional Order Proportional-Integral-Derivative (FOPID) controllers to achieve optimal speed regulation of DC motors. The LHHO algorithm is an innovative meta-heuristic algorithm that draw inspiration from the cooperative hunting behavior and leadership prowess of the Harris Hawks called the “seven pounds”. While several error functions were tested in this study, the integral of time multiplied absolute error (ITAE) has been adopted as the error function for obtaining the parameters of PID and FOPID controllers using the LHHO algorithm. Through quantitative evaluations and comparisons with existing techniques, the proposed controllers (LHHO-PID and LHHO-FOPID) have revealed significant improvements in key performance metrics, including rise time, settling time, and maximum overshoot during transient periods when ITAE is used as the error function. Furthermore, the stability response and robustness analyses were carried out by varying the parameters of the DC motor under eight scenarios, which confirmed competitive performance in system response and transient behavior.

Real-Time Implementation of BLDC Motor-Based Intelligent Tracking Control Fed from PV-Array for E-Bike Applications

The essential goal of this research is designing and modeling a speed and position tracking system for driving an electric bike (e-bike) motordrive. This motor is a brushless DC (BLDC) motor as a high-performance drive. It is supplied from twin electric sources to drive it and charge the storage elements (i.e., batteries, super-capacitors, etc.). The first one is a renewable, neat, and clean source photovoltaic (PV) module and the second one is a pedal generator driven by the rider. The submitted design of the controllers is optimized to improve the system’s dynamic stability. The artificial bee colony (ABC) as an artificial intelligent (AI) algorithm is suggested for searching the optimal gains of the proposed proportional-integral-derivative (PID) controllers by reducing the error of its fitness function. The system behavior is studied with that controller when directly feeding from the PV array with and without batteries. The response of the proposed technique - against dynamic troubles and PV oscillations such as irradiance- is also verified. Other evolutionary computational techniques - such as ant colony optimization (ACO) and genetic algorithm (GA)- have been compared with the behavior of the proposed controller to ensure high efficiency in optimized tuning of PID gains. Then, the proposed controller that gives a high performance will be executed in real-time by using OPAL-RT 4510 RT-simulator and rapid control prototyping.

Performance Analysis of Sensor less Induction Motor Drive using Improved Control Techniques

AC Drives demand robust motor design with rugged construction, low cost, high reliability in service, and simple maintenance. In modern power drives, Sensorless Induction motor drives are more popular than other drives. A speed sensor/encoder-based drive is costlier and requires more space in case more parallel units are coupled together in drive operation. To address these difficulties, speed sensorless drives are introduced without loss of efficiency and reliability. However, Sensorless speed drive requires advanced control techniques in which complex calculations are there due to the nonlinearity of IM. In the recent past, Machine model-based methods i.e., MRAS (Model reference adaptive systems) and ELO (Extended Luenberger observer) have been popularized due to ease of implementation compared to other techniques. In this work, improved MRAS and ELO models are implemented and verified using Simulink software. Then, the same is validated with the help of an FPGA (field programmable gate array) based Snetly real-time controller. The key improvements are achieved as follows: effective speed estimation, robust speed tracking, reduced speed estimation error, and robust stability which helps to enhance the transient and steady-state performance of the drive.