Search

With the widespread application of rehabilitation robots, surface electromyography (sEMG), as a non-invasive way, has become an effective tool to judge the activation state of the related muscles. The traditional muscle activation model is not accurate enough for the reason that the sEMG signals contain wealth of unknown information. To estimate the muscle activation state and to evaluate the effect of ankle rehabilitation training, the activation mean sample entropy is proposed in this paper which is modified by the sample entropy of the complexity of the expression sequence based on the traditional muscle activation model. The degree of muscle activation in the process of ankle rehabilitation training was evaluated by using a parallel ankle robotic system, and the accuracy of the index was verified by conducting experiments with a healthy subject.

Because of low efficiency and accuracy in the domain in the’ oposed an evaluation model based on the method of deep learning and designed a optimal RBM training algorithm. The comparative experimental model in this paper mainly adopts naive Bayesian model algorithm, restricted Boltzmann machine model algorithm and perceptron model algorithm, respectively based on the training data of the algorithm model in this paper. The algorithm proposed in this paper can obtain high calculation accuracy under the condition of given training sample data, and also has good calculation accuracy in terms of running time. In the process of model training, the model proposed in this paper can obtain high accuracy in less training time, and the model has the least calculation time. The results have a good capacity of extracting features and can reflect the essential features by comparing the result with other different modern optimization algorithms.

In this work, we approach high-precision attitude control problem of UAV manipulators in presence of external disturbances. A disturbance observer is exploited to estimate the disturbance, meanwhile we used the estimated result as a feed-forward compensation. A backstepping controller is employed to stabilize the UAV attitude by feedback. The mathematical model of UAV manipulator is constructed by combining the UAV base and the equipped robotic arm together. Furthermore, to relieve the influence of sensor noise, we fused a tracking differentiator into the whole system. The constructed method is validated in realistic simulations, where we executed UAV manipulator hovering flight with varying external disturbances.

The super-twisting algorithm has been verified its effectiveness in attitude control for quadrotor with unknown disturbance. However, the discontinuous signum function in super-twisting formulation causes the unwanted chattering problem and thus affects the attitude control performance. In this paper, we replace the signum function with sigmoid function and propose the sigmoid super-twisting extended state observer (SSTESO) and sliding mode controller (SSTSMC). Besides, a rigorous convergence analysis of proposed nonlinear error system based on the Lyapunov stability theory is performed, and the parameter selection rule is given. Numerical simulations are implemented in presence of manually designed disturbance in order to verify the advantage of the proposed algorithm compared with original super-twisting algorithm. Results show that the introduction of sigmoid function can not only reduce the chattering effect, but also maintain the advantage of supertwisting algorithm.

In this paper, we are analyzed the switching characteristics of SiC MOSFET. According to the waveform of gate voltage VGS, drain voltage VDS and drain current ID of SiC MOSFET in switching process, the whole turn-on/off process can be divided into four phase: phase I, turn-on/off delay. Phase II, current/voltage rise. Phase III, voltage/current fall. Phase IV, VGS rise to VGG /0. The voltage and current changes at each phase are analyzed. To directly observe the parameters change, we draw a turn-on/off process diagram according to Sic MOSFET characteristics. As temperature variety, turn-on/off process are significantly affected and parameters change rate is proportional to the temperature.

This paper presents the implementation of an autonomous aerial manipulation using a hexacopter equipped with a two DOF robotic arm. The kinematic and dynamic models are developed by considering the dynamic characteristics of the combined manipulation platform. A novel adaptive sliding mode controller is proposed for both position and velocity control. By building SSD (Single Shot Detection) deep neural network based on deep learning, an object detection solution is developed. The three dimensional coordinates of the target object relative to the multirotor are obtained via combining the vertical plane position obtained from object detection law with the depth value from stereo camera. Finally, with the proposed controller and object detection law, an autonomous flight experiment is accomplished including approaching, grabbing and delivering the target object. The proposed approaches are demonstrated with effectiveness and could be utilized in various manipulation applications.

To overcome the shortcomings of the centralized control method, such as large calculation, strong communication dependency and poor system reliability. First, based on the traditional IEEE14 standard network topology, this paper introduces the controllable micro-source consist of PV and energy storage basic units. Micro-source cost model and constraints of photovoltaic renewable power generation unit, controllable flexible load, energy storage unit and so on are established. Next, consensus algorithm with no leader of two different communication coefficients D and M, N (CANLD and CANLMN) are finished when controllable micro-sources are inserted into micro-grid, and finally the eigenvalue perturbation method is used to prove the convergence of two different consensus algorithms. The results of the case study show that the two different multi-agent non-leader algorithms with traditional energy and controllable microsources incremental cost as the consistency variable can satisfy the optimal economic objective in microgrid and CANLMN can reach the incremental cost consistency of each micro-source with smaller computation and faster speed.

The quadrotor equipped with a robotic arm has been applied to many fields recently. This paper deals with the problem of quadrotor stabilization and trajectory tracking equipped with a 2-DOF robotic arm. By considering the robotic arm as an independent part to affect the center of gravity (COG) of the vehicle, the kinematic and dynamic models are derived. The controller for the whole system is designed with two stages. First, a backstepping sliding mode controller is applied to position tracking (outer loop) and generating desired attitudes for inner loop. Then a backstepping controller with integral action is built for attitude stabilization (inner loop). In addition, a COG compensation system is developed for the controller to cope with the COG variation caused by movement of the robotic arm. Finally, several simulations demonstrate the effectiveness of the proposed control approaches.

In this paper, we present a partial steganography algorithm for 3D triangle mesh. Traditional 3D steganography hides information among the whole mesh and our method random picks some vertices for embedding. The random picking processing works with the pseudorandom number generator (PRNG) and a secret seed for the PRNG. The same random sequence can be disinterred with the same seed so that we can extraction the secret message from the stego model. In experiments, our method is effective against the 3D Steganalysis as best 15.2% reduction in classification accuracy of a supervised classification.

This paper proposes an adaptive robust control (ARC) scheme for quad-rotor to track 3D trajectory subject to unknown external disturbances and parametric uncertainties. The proposed controller is based on a compound control method including back-stepping and sliding mode control (SMC). Back-stepping is implemented to handle the nonlinearity of the dynamic model, complementary SMC is used to compensate the effect of external disturbances, and adaptive control is applied to estimate system parameters like mass in real time. The asymptotic stability of the proposed ARC scheme has been proved by the Lyapunov stability theory. Numerical simulations in various cases show the effectiveness, adaptability and robustness of the proposed ARC compared with PID control and robust control (RC) without adaptive control law.

EV (Electric vehicle) and PV (photovoltaic) load will bring about some influence to the power system when is connected to gird, such as the change of load characteristics, increase the peak valley difference and etc. At the same time the application of inverter will bring about reverse power threat to the operation and maintenance of power grid. To weaken the influence above, the technology named Vehicle to Home is proposed, which can control EV charging load to a specific value as well as control the EV discharge to grid. In order to actively respond to the impact of EV and distributed PV application access to the power grid, based on the actual home users, the data of household electricity load and the user level transformer are collected and analyzed. At the same time with the peak and valley price, based on the quantitative analysis of the influence of electric cars, solar access to the grid, the interactive applications is put forward. Calculation and interactive analysis results show that the implementation of V2H can not only bring economic benefits to users but also reduce the peak valley ratio and the risk of peak on peak which result on the electric vehicle charging in disorder.