Search results

Multi-domain convolutional neural network (MDCNN) model has been widely used in object recognition and tracking in the field of computer vision. However, if the objects to be tracked move rapid or the appearances of moving objects vary dramatically, the conventional MDCNN model will suffer from the model drift problem. To solve such problem in tracking rapid objects under limiting environment for MDCNN model, this paper proposed an auto-attentional mechanism-based MDCNN (AA-MDCNN) model for the rapid moving and changing objects tracking under limiting environment.

First, to distinguish the foreground object between background and other similar objects, the auto-attentional mechanism is used to selectively aggregate the weighted summation of all feature maps to make the similar features related to each other. Then, the bidirectional gated recurrent unit (Bi-GRU) architecture is used to integrate all the feature maps to selectively emphasize the importance of the correlated feature maps. Finally, the final feature map is obtained by fusion the above two feature maps for object tracking. In addition, a composite loss function is constructed to solve the similar but different attribute sequences tracking using conventional MDCNN model.

In order to validate the effectiveness and feasibility of the proposed AA-MDCNN model, this paper used ImageNet-Vid dataset to train the object tracking model, and the OTB-50 dataset is used to validate the AA-MDCNN tracking model. Experimental results have shown that the augmentation of auto-attentional mechanism will improve the accuracy rate 2.75% and success rate 2.41%, respectively. In addition, the authors also selected six complex tracking scenarios in OTB-50 dataset; over eleven attributes have been validated that the proposed AA-MDCNN model outperformed than the comparative models over nine attributes. In addition, except for the scenario of multi-objects moving with each other, the proposed AA-MDCNN model solved the majority rapid moving objects tracking scenarios and outperformed than the comparative models on such complex scenarios.

This paper introduced the auto-attentional mechanism into MDCNN model and adopted Bi-GRU architecture to extract key features. By using the proposed AA-MDCNN model, rapid object tracking under complex background, motion blur and occlusion objects has better effect, and such model is expected to be further applied to the rapid object tracking in the real world.

This paper aims to focus on the tracking of a moving magnetic target by using total field magnetometers and to present a tracking method based on the gradient of a magnetic anomaly. In the tracking, it is assumed that the motion of the target is equivalent to a first-order Markov process. And the unit direction vector of the magnetic moment from the gradient of the magnetic anomaly can be obtained. According to the unit direction vector, the inverse problem is turned into an optimization problem to estimate the parameters of the target. The particle swarm optimization algorithm is used to solve this optimization problem. The proposed method is validated by the numerical simulation and real data. The parameters of the target can be calculated rapidly using the proposed method. And the results show that the estimated parameters of the mobile target using the proposed method are very close to the true values.

The authors focus on the tracking of a moving magnetic target by using total field magnetometers and present a tracking method based on the gradient of a magnetic anomaly.

The paper provides an effective method for tracking the magnetic target based on an array with total field sensors.

Comparing with a vector magnetic sensor, the measurement of the scalar magnetic sensor is almost not influenced by its orientation. In this paper, a moving magnetic target was tracked by using total field magnetometers and a tracking method presented based on the gradient of a magnetic anomaly.

This study aims to investigate the rule‐based decentralised control framework for a swarm of UAVs carrying out a cooperative ground target engagement mission scenario.

This study is to investigate the rule‐based decentralised control framework for missions which require high‐level cooperation between team members. The design of the authors’ control strategy is based on agent‐level interactions. Different to a centralized task assignment algorithm, the cooperation of the agents is entirely implicit. The behaviour of the UAVs is governed by rule sets which ultimately lead to cooperation at a system level. The information theoretic measures are adopted to estimate the value of possible future actions. The prediction model is further considered to enhance the team performance in the scenario where there are tight coupled task constraints.

The simulation study evaluates the performance of the decentralised controller and compares it with a centralised controller quantitatively. The results show that the proposed approach leads to a highly cooperative performance of the group without the need for a centralised control authority. The performance of the decentralised control depends on the complexity of the coupled task constraints. It can be improved by using a prediction model to provide information such as the intentions of the neighbours that is not available locally.

The achievable performance of the decentralised control was considered to be low due to the absence of communication and little global coordinating information. This study demonstrated that the decentralised control can achieve highly cooperative performance. The achievable performance is related to the complexity of the coupled constraints and the accuracy of the prediction model.

The purpose of this paper is to take transient contact force response, overshoots and steady-state force tracking error problems into account to form an excellent force controller.

The basic impedance function with a pre-PID tuner is designed to improve the force response. A dynamic adaptive adjustment function that combines the advantages of hybrid impedance and adaptive hybrid impedance control is presented to achieve both force overshoots suppressing and tracking ability.

The introduced pre-PID tuner impedance function can achieve more than the pure impedance function in aspects of converging to the desired value and reducing the force overshoots. The performance of force overshoots suppression and force tracking error are maintained by introducing the dynamic adaptive sigma adjustment function. The simulation and experimental results both show the achieved control performance by comparing with the previous control methods.

The implementation of the controller is easy and convenient in practical manufacture scenes that require force control using industrial robots.

A superior robot controller adapting to a variety of complex tasks owing to the following characteristics: maintenance of high-accuracy position tracking capability in free-space (basic capabilities of modern industrial robots); maintenance of high speed, stability and smooth contact performance in collision stage; and presentation of high-precision force tracking capability in steady contact.

This study aims to deal with the problem of trajectory tracking control for the quadrotor under external environmental disturbance and variable payloads.

In the field of unmanned aerial vehicle (UAV) control, external environmental disturbance and internal variable payloads as two major interference factors lead to control performance degradation or even instability, thus a trajectory tracking controller which innovatively combines sliding mode control technology and model-free control technique is proposed. The proposed controller is constructed with a learning rate-based sliding mode controller and an ultra-local model. Based on the proposed controller, the nonlinear system model of variable load quadrotor is locally estimated and the system’s uncertainties and disturbances can be compensated.

The simulation and actual test results demonstrate the satisfactory control performance and the robustness of the proposed controller compared with the PID and Backstepping controller under external environmental disturbance and variable payloads. Moreover, the proposed controller solves the trajectory tracking control problem not only when payloads change at the center of gravity but also when the position of load variation deviates from the center of gravity.

In both military and civilian domains, the quadrotor may encounter such situations that the payloads change, such as transporting goods, aerial refueling and so on. As a large internal interference factor, variable load tends to lead to unstable control. The research results provide theoretical guidance and technical support for trajectory tracking control of quadrotor under variable payloads.

The proposed controller combines learning rate-based sliding mode controller and model-free control technique to achieve a more efficient and accurate trajectory control of the quadrotor when considering system uncertainties and the load variation that happens in the unknown location.

The purpose of this paper is to focus on perfect trajectory tracking control of 2 DOF non‐holonomic mobile robots in which the guidance and control commands are imposed through independent driver wheels. Model‐based nonlinear controllers for these robots with unknown parameters require estimation of a specified set of the robot parameters. The effects of the proposed model dynamics in both local and global coordinate systems are fully examined on the parameter estimation and tracking performance.

Design of suitable feedback linearization (FL) controllers for trajectory tracking control of wheeled mobile robots (WMRs) is first reviewed. A FL controller whose parameters are tuned using fuzzy computations (fuzzy if‐then rules) is then developed. In the line of the other contributions of the paper, a pure fuzzy controller that is merely based on fuzzy if‐then rules is proposed to trajectory tracking control of the mobile robots.

Use of global dynamics for designing a suitable FL control system leads to a perfect compensation for initial off‐tracks. Furthermore, the estimated parameters are unbiased because the corresponding regressor/signal matrix indicates a high rank of persistent excitation. Fuzzy tuning of the controller instead of keeping the gains fixed makes the overall system more robust against measurement noises while upper bounds and fluctuations of the input torques are both remarkably reduced. The pure fuzzy controller is naturally independent of the robot dynamics and therefore, the necessity of parameter estimation algorithm is removed.

The paper provides some new nonlinear controllers for WMRs, in order to make perfect trajectory tracking and initial off‐tracks compensation.

Considering the unmeasurable states of the systems and the previewed reference signal, a novel fuzzy observer-based preview controller, which is a mixed controller of the fuzzy observer-based controller, fuzzy integrator and preview controller, is considered to address the tracking control problem.

The authors employ an augmentation technique to construct an augmented error system for uncertain T-S fuzzy discrete-time systems with time-varying uncertainties. Additionally, the authors obtain the corresponding linear matrix inequality (LMI) conditions for designing the preview controller.

This paper discusses the preview tracking problem for nonlinear systems. First, considering the unmeasurable states of the systems and the previewed reference signal, a novel fuzzy observer-based preview controller, which is a mixed controller of the fuzzy observer-based controller, fuzzy integrator, and preview controller, is considered to address the tracking control problem. Then, using the fuzzy Lyapunov functional with the linear matrix inequality (LMI) technique, new sufficient conditions for the asymptotic stability of the augmented system are derived by applying the LMI technique. The preview controller and fuzzy observer can be designed in one step. Finally, a numerical example is used to illustrate the effectiveness of the results.

An augmented error system is successfully constructed by the state augmentation approach. A novel preview controller is designed to address the tracking control problem. The preview controller and fuzzy observer can be designed in one step.

The purpose of this paper is to design a global robust and continuous control scheme for the attitude tracking control problem of the reentry vehicle with parameter uncertainties and disturbances.

First, feedback linearization is applied to the model of reentry vehicle, resulting in three independent uncertain subsystems. Then a new second-order time-varying sliding function is proposed, based on which a continuous second-order time-varying sliding mode control (SOTVSMC) law is proposed for each subsystem. The global robustness and convergence performance of the closed-loop reentry vehicle control system under the proposed control law are proved.

Simulation is made for a reentry vehicle through the assumption that there is external disturbance to aerodynamic moment and the aerodynamic parameters as well as the atmospheric density are perturbed. The results verify the validity and robustness of the proposed strategy.

The SOTVSMC attitude controller based on feedback linearization is proposed for the reentry vehicle. The advantages of the proposed SOTVSMC are twofold. First, the global second order sliding mode is established, which implies that the closed-loop system is global robust against matched parameter uncertainties and disturbances in reentry. Second, the chattering problem is significantly alleviated.

This paper aims to introduce a new target tracking method based on risk theory in a 2-D discrete environment. After that, the related sensor scheduling method is proposed. This can make up the blank of target tracking and sensor management in the 2-D discrete environment.

The definition of risk is proposed based on risk decision theory firstly. Then the target tracking model in a two-dimensional discrete environment is built. The motion state updating and estimation method of target’s motion state based on Bayes theory is given. Thirdly, the method of computing sensor emission interception risk is provided. Afterwards, the optimization rule of obtaining the minimum risk is followed to model the sensor scheduling objective function. The lion algorithm is adjusted and improved combined with Chaos theory to generate the optimal sensor management projects.

The risk-based sensor target tracking method and sensor management method are both effective in a 2-D discrete environment.

To the best of the authors’ knowledge, this paper is the first to study the target tracking method and sensor scheduling method in a 2-D environment. Furthermore, the lion algorithm is improved combined with Chaos theory to show a better optimization performance.

The research of unmanned air/ground vehicle (UAV/UGV) cooperation has attracted much attention due to its potential applications in disaster rescue and target surveillance. This paper aims to focus on the UAV/UGV cooperative target tracking and enclosing, considering the limits of detection and sensor failures.

The UAV/UGV cooperation structure is designed, contributing to homogeneous consistency and heterogeneous communication. The target tracking of UAVs is converted into a constraint optimization problem involving tracking cost, and the target enclosing of UGVs is modeled as formation control.

The energy estimation pigeon-inspired optimization is developed to generate control inputs for UAVs. And the controller combined with switchable topology is proposed, where the switching rule is flexible in dealing with some emergencies.

The proposed structure and algorithms can be easily applied to practice and help design the UAV/UGV control system.

The energy estimation mechanism is proposed for the target tracking of UAVs, and the rules of switching topologies ensure the target enclosing process of UGVs.