A nonlinear system model for a class of hydraulic device was constructed.On the basis of the conversion of its model,an exponent approaching law with variable parameter was adopted to realize the sliding mode control. A position tracking with high precision was realized,and it had a better disturbance-resistant ability . A boundary layer technique was used to reduce the chattering in sliding mode control .The simulation result proves that the designed controller can achieve the scheduled demand and has a good effect.
Abstract A hierarchical controller is proposed for achieving high‐accuracy control and the dynamic balance with the presence of multiple faults of actuator, the external disturbance, and the model uncertainties in multicylinder hydraulic press machine (MCHPM). The method divides the controller design into three steps: Virtual fault‐tolerant control law, control allocation algorithm, and actuator control law, which are progressive. First, to precisely compensate the lumped disturbances including the multiple faults of actuator, the external disturbance, and the model uncertainties, a disturbance observer (DO) is developed. By combining the observer with the sliding mode control (SMC), a virtual fault‐tolerant control law is designed. Second, a highly integrated control allocation algorithm for the virtual fault‐tolerant control law is proposed to get the desired driving force, taking into account dynamic control allocation (DCA), multiobjective optimization (MOO) and Analytic Hierarchy Process (AHP) simultaneously. Third, taking the driving force obtained from above control allocation algorithm as the desired target, the control law of each cylinder is calculated. The global stability for the whole system is proved by the Lyapunov theory. Lastly, results of simulation and experiment show that the proposed controller can effectively handle different faults and have more superior control performance.
CANopen is a kind of higher layer protocol based on CAN, and it is widely used in industrial automation field, especially in distributed motion control systems. The paper presents an approach of realizing CANopen slave node. Microcontroller STC90C514RD is used as a core chip to design the control system; the design adopts CTM1050 as the CAN transceiver and uses SJA1000 as the CAN controller. And the system consists of a CANopen slave node, input module and output module. MicroCANopen, which is an open source protocol stack of CANopen, is transplanted to main controller so that the function of CANopen could be implemented. Moreover the experimental platform is built to test the design of the system. The results prove that the design is feasible and valid.
According to the principle of forecast feed forward compensation dynamic matrix control (DMC) and regarding its low calculating accuracy, bidirectional decoupling DMC is presented and applied in random long delay multi-input and multi- output (MIMO) networked control system (NCS) in this paper. This algorithm, therefore, expands the using field of DMC and resolves control problems in random long delay NCS. The feasibility and validity were finally demonstrated by the results of simulating experiments.
Abstract Considering the load uncertainty and unmodeled dynamics in multicylinder hydraulic systems, this paper proposes a balance control algorithm based on safe reinforcement learning to release the restrictions of classical model-based control methods that depend on fixed gain. In this paper, the hydraulic press is controlled by a trained agent that directly maps the system states to control commands in an end-to-end manner. By introducing an action modifier into the algorithm, the system states are kept within security constraints from the beginning of training, making safe exploration possible. Furthermore, a normalized exponential reward function has been proposed. Compared with a quadratic reward function, the precision is greatly improved under the same training steps. The experiment shows that our algorithm can achieve high precision and fast balance for multicylinder hydraulic presses while being highly robust. To the best of our knowledge, this research is the first to attempt the application of a reinforcement learning algorithm to multi-execution units of hydraulic systems.
This study addresses the tracking–learning–detection (TLD) algorithm for long‐term single‐target tracking of moving vehicle from video streams. The problems leading to tracking failures in existing TLD methods are discovered, and an improved TLD (ITLD) tracking algorithm is proposed which is more robust to object occlusion and illumination variation. A square root cubature Kalman filter (SRCKF) is employed in the tracker of TLD to predict the position of the object when occlusion occurs. Besides, this study introduces fast retina keypoint (FREAK) feature into the tracker to alleviate the instability caused by illumination variation or scale variation. The overlap comparison and the normalised cross‐correlation coefficient (NCC) are introduced to the integrator of the TLD to obtain reliable bounding boxes with improved tracking precision. Experiments are conducted to compare the performance of the state‐of‐the‐art trackers and the proposed method, using the object tracking benchmark that includes 50 video sequences (OTB‐50) and TLD datasets. The experimental results show that the proposed ITLD outperforms on both tracking accuracy and robustness. The proposed method can track a moving vehicle even when it is temporally totally occluded.
Sliding mode variable structure control theory is a kind of control method deriving from the development of computer. It has unique and excellent robustness to the system parameter perturbation, external disturbance, system uncertainty and so on, and the algorithm is simple. In this thesis, compared with the traditional equal velocity trending law and traditional exponential velocity trending law, the chattering problem of sliding mode variable structure control system is improved. The results show that the advantages of the improved exponential velocity trending law in preventing and eliminating chatter are summarized, and a feasible control scheme is proposed, which has achieved good results in solving the chattering problem of variable structure control systems. It provides a feasible solution for simply and quickly solving system stability problems and improving system performance.
In this paper, a multi-agent–based reinforcement learning (RL) algorithm is proposed to solve the leveling control problem of a multi-cylinder hydraulic press with coupling phenomena. This algorithm is a model-free control algorithm, which can avoid the modeling difficulties and low efficiency caused by the complexity of the model. The control algorithm of the hydraulic press adopts Multi-Agent Soft Actor–Critic (MASAC). The concept of multi-agent is introduced to control each coupling input separately. The distributed updating method is used to realize accurate and stable control of the hydraulic press. At the same time, a reward function of the piecewise function type is proposed in this paper. Compared with common algorithms such as the quadratic reward function, this algorithm has a faster and more stable convergence effect in the whole process. Experiments show that the proposed algorithm has better convergence speed and leveling accuracy than the traditional single-agent algorithm.
Deep neural network is a hotspot in the field of Machine Learning, which can realize deep hierarchical representation of input data. In this paper, a simplified Shallow Convolutional Neural Network (SCNN) is used to classify Motor Imagery Electroencephalogram (MI-EEG). The network uses a single feature layer to mine the intrinsic features of EEG sequences, and uses the training time and accuracy to evaluate the classification effect. Different from other methods, CNN integrates feature extraction and predicted classification results to optimize the whole process, which can automatically extract features from original data. Experiments on the BCI Competition IV Dataset 2b competition dataset and the BCI Competition II Dataset III competition dataset show that the overall performance under SCNN is the best.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
