In this paper, a novel control method of the redundant force branch based on the force/position hybrid control structure of Smith predictor compensation is proposed. A fuzzy PI controller is designed based on Smith pr...
详细信息
In this paper, a novel control method of the redundant force branch based on the force/position hybrid control structure of Smith predictor compensation is proposed. A fuzzy PI controller is designed based on Smith predictor compensation structure and it is included in the redundant force branch. This method can obtain good tracking and dynamic performance. However, fuzzy control doesn't have self-learning and adaptive ability, so fuzzy neural network (FNN) controller is used in the redundant force branch. The simulation results show that the proposed FNN algorithm based on delay compensation force/position hybrid control structure can improve the adaptability and the control accuracy of driving force of redundant branch.
In this paper, a new concept, the fuzzy rate of an operator in linear spaces is proposed for the very first time. Some properties and basic principles of it are studied. Fuzzy rate of an operator B which is specific i...
详细信息
This paper proposes a system, by which humanoid robots could imitate human movements to learn human behaviors. Considering the security problem during humanoid robot imitation process, a method of stability identifica...
详细信息
This paper proposes a system, by which humanoid robots could imitate human movements to learn human behaviors. Considering the security problem during humanoid robot imitation process, a method of stability identification based on improved Multi-Layer Perception (MLP) is proposed. Firstly, the positions of the human joints are collected by a Kinect. Secondly, the human joints information is transformed into robot angles information by the inverse kinematics equation. Finally, the stability of the robot is identified by the deep neural network, which has the robot angles as its input. Simulations and experimental results demonstrate that compared with the Back Propagation neural network (BPnn), RBF neural network (RBFnn), Support Vector Machine (SVM) and Extreme Learning Machine (ELM), MLP neural network has the best identification accuracy. And the filters we used can overcome the problem of speed pulse, to some extent, and achieved an ideal imitation result.
Compound wave defect occurs in flatness during tandem cold rolling when high-end automobile and household appliance plate are produced, which has several impact on line speed and quality of final product. In this pape...
详细信息
To solve the problem of premature convergence and search stagnation in traditional differential algorithm, this paper presents an adaptive dual model differential evolution algorithm based on cloud model(ADDEC). The v...
详细信息
To solve the problem of premature convergence and search stagnation in traditional differential algorithm, this paper presents an adaptive dual model differential evolution algorithm based on cloud model(ADDEC). The variation model of ADDEC is produced by combining basic variation with cloud variation. And the adaptive operators are referenced respectively for scale factor and interactive probability during the operation of variation and crossover. The robustness and the convergence rate of the algorithm are enhanced. Moreover the relationship between global search and local search of the algorithm is improved. The diversity of the population is guaranteed at the same time. It is tested by typical high dimension benchmark function and compared with other algorithms, and the results show that the ADDEC has better precision, high convergence speed and makes it easy to jump out the local optimal solution.
This paper deals with the formation control problem for a group of quadrotor unmanned aerial vehicles(UAVs) through a modified dynamic event-triggered control protocol. The stability proof for the state space origin...
详细信息
This paper deals with the formation control problem for a group of quadrotor unmanned aerial vehicles(UAVs) through a modified dynamic event-triggered control protocol. The stability proof for the state space origin of the overall closed-loop system is derived from the singular perturbation method and Lyapunov stability theory. An event-triggered formation control protocol in terms of a dynamically varying threshold parameter is delicately carried out while acquiring satisfactory resource efficiency, and Zeno behavior of triggering time sequences is excluded. Finally, simulations on four quadrotor UAVs are given to verify the effectiveness of the theoretical results.
This paper proposes a fuzzy Q-learning (FQL) algorithm to solve the problem of the robot obstacle avoidance in unknown environment. FastSLAM algorithm is used to localize the position of the robot. Traditional Q-learn...
详细信息
To improve the global search ability under the condition of ensuring convergence speed, it is still a major challenge for most meta-heuristic optimization algorithms. The Moth-Flame Optimization (MFO) algorithm is an ...
详细信息
To improve the global search ability under the condition of ensuring convergence speed, it is still a major challenge for most meta-heuristic optimization algorithms. The Moth-Flame Optimization (MFO) algorithm is an innovative nature-inspired algorithm. To improve the precision of the solution and to quicken the convergence speed and to increase the stability of MFO, an ameliorated Moth-flame optimization algorithm (A-MFO) that combines the crisscross optimization algorithm with MFO is proposed to solve this problems that are mentioned above. The performance of proposed A-MFO is demonstrated on six benchmark mathematical function optimization problems regarding superior accuracy and lower computational time achieved compared to other well-known nature-inspired algorithms.
The problem of sensor fault diagnosis and reconstruction of the continuous casting mold vibration displacement system driven by servo motor is investigated in this paper, and a fault reconstruction method based on ada...
详细信息
The problem of sensor fault diagnosis and reconstruction of the continuous casting mold vibration displacement system driven by servo motor is investigated in this paper, and a fault reconstruction method based on adaptive sliding mode observer is proposed. Firstly, the matching condition of the observer is satisfied by constructing an auxiliary output, and a new state variable is selected as the low-pass filter with the fault vector output. Then an adaptive sliding mode observer is designed for the auxiliary output system in the present of the unknown upper bound of the fault, in which the adaptive switching gain is updated online. The gain matrix of the observer is obtained by solving a LMI optimization problem. And a linear transforming matrix is introduced to realize the direct estimation of the fault by the principle of the equivalent control output error injection. Finally, simulation results are given to illustrate the effectiveness of the proposed method.
Comprehensive learning particle swarm optimization(CLPSO) algorithm has a good performance in overcoming premature convergence and avoiding getting stuck in local minima, which are shortcomings in particle swarm opt...
详细信息
ISBN:
(纸本)9781538629185
Comprehensive learning particle swarm optimization(CLPSO) algorithm has a good performance in overcoming premature convergence and avoiding getting stuck in local minima, which are shortcomings in particle swarm optimization. It can solve complex, multi-modal of single-objective problems, but it has not such performance in handling multi-objective optimization problems because of the difficulty of selective solution mechanism. In this article, a multi-objective decomposition particle swarm optimization based on comprehensive learning strategy is proposed, which uses a comprehensive learning strategy for multi-objective problems to prevent premature convergence;updates the leading particles by decomposition method to enhance the distribution of solutions;adds the archive to preserve non-dominated solutions, and adopts mutation in archive to avoid falling into local optimum. The proposed approach is compared with three multi-objective evolutionary algorithms and the results indicate that the proposed approach is competitive respect to which it is compared in most of the test problems adopted.
暂无评论