Robots as therapy tools have been researched in intervention for children with autism. During the interaction between robots and autistic children, engagement is an important metric which can be used to express whether robot's behavior is suited to the current context. The evaluation of engagement is a key prerequisite to improve the autonomous ability of robots in intervention. In this paper, we propose a new model to evaluate the engagement of children with autism. The proposed model is developed based on the dynamic Bayesian network, and the parameters of the model are obtained by fuzzy logic and expert elicitation. After determining the input features and the classification of engagement, the topology of the model is established. Afterward, experts' opinions are collected based on linguistic variables. Based on triangular fuzzy number, the parameterization of the model is realized by fuzzification, aggregation, and defuzzification. Finally, the model is validated by experiment. The result demonstrates that proposed model satisfies the actual demands and the result of engagement classification can provide the input condition for the decision making of the robot.

During target capturing operation, the changes in the dynamics parameters of a free-floating space manipulator degrade the performance of the base attitude stabilization. This paper presents a new self-learning soft-grasp control algorithm based on the variable stiffness technology. First, the dynamic model of variable stiffness joint space manipulator system is established. Simultaneously, the detailed dynamic analysis of pre-impact and post-impact stages is carried out. Second, a new soft-grasp control strategy utilizing cellular differential evolution algorithm combined opposition-based learning with orthogonal crossover is employed to minimize the base angular momentum. Its principle is to solve the optimal stiffness value of the variable stiffness joint to realize desired buffering. Thereafter, we put forward an adaptive backstepping sliding mode control method to track the actual joint stiffness. Finally, the proposed method is applied to a two-degree of freedom planar free-floating space manipulator and the simulation results verify the effectiveness.

The limited number of brain–computer interface based on motor imagery (MI-BCI) instruction sets for different movements of single limbs makes it difficult to meet practical application requirements. Therefore, designing a single-limb, multi-category motor imagery (MI) paradigm and effectively decoding it is one of the important research directions in the future development of MI-BCI. Furthermore, one of the major challenges in MI-BCI is the difficulty of classifying brain activity across different individuals. In this article, the transfer data learning network (TDLNet) is proposed to achieve the cross-subject intention recognition for multiclass upper limb motor imagery. In TDLNet, the Transfer Data Module (TDM) is used to process cross-subject electroencephalogram (EEG) signals in groups and then fuse cross-subject channel features through two one-dimensional convolutions. The Residual Attention Mechanism Module (RAMM) assigns weights to each EEG signal channel and dynamically focuses on the EEG signal channels most relevant to a specific task. Additionally, a feature visualization algorithm based on occlusion signal frequency is proposed to qualitatively analyze the proposed TDLNet. The experimental results show that TDLNet achieves the best classification results on two datasets compared to CNN-based reference methods and transfer learning method. In the 6-class scenario, TDLNet obtained an accuracy of 65%±0.05 on the UML6 dataset and 63%±0.06 on the GRAZ dataset. The visualization results demonstrate that the proposed framework can produce distinct classifier patterns for multiple categories of upper limb motor imagery through signals of different frequencies. The ULM6 dataset is available at https://dx.doi.org/10.21227/8qw6-f578.

Aiming at the soft contact problem of space docking, a bionic docking mechanism for space target acquisition is proposed to realize the buffering and unloading of six–dimensional spatial collision through flexible rotating and linear components. Using the Kane method, an integrated dynamic equation of the bionic docking mechanism in space docking is established, and the stiffness optimization strategy is carried out based on angular momentum conservation. Based on the particle swarm optimization (PSO), a stiffness optimization scheme was realized. Through the numerical simulation of the bionic docking mechanism in space docking, the stiffness optimization was achieved and the soft contact machine process is verified. Finally, through the docking collision experiments in Adams, the results indicate that the proposed bionic docking mechanism can not only prolong the collision time to win time for space acquisition, but also buffer and unload the six–dimensional spatial collision caused by space target docking.

This paper presents a new global terminal sliding mode control (GTSMC) methodology for the simultaneous positioning and vibration suppression of slewing a flexible-link manipulator. The proposed control scheme is realized using only a joint actuator without an extra actuator (for example, the widely used piezoelectric zirconate titanate), and the non-minimum phase vibration control problem is solved. Firstly, based on the differential geometry method, the initial dynamic system is decomposed into two subsystems, namely an input-output subsystem and a zero-dynamics subsystem. Secondly, a continuous GTSMC strategy without chattering phenomenon is designed for the input-output subsystem, and the limited convergence time is deduced. Moreover, the eigenvalues of the zero-dynamics subsystem can be pointed by setting proper controller parameters for Lyapunov asymptotical stability. Finally, simulation and experimental results demonstrated the efficacy and feasibility of the proposed control methodology.