Upper extremity exoskeleton rehabilitation robots have become a significant piece of rehabilitation equipment, and planning their motion trajectories is essential in patient rehabilitation. In this paper, a multistrategy improved whale optimization algorithm (MWOA) is proposed for trajectory planning of upper extremity exoskeleton rehabilitation robots with emphasis on isokinetic rehabilitation. First, a piecewise polynomial was used to construct a rough trajectory. To make the trajectory conform to human-like movement, a whale optimization algorithm (WOA) was employed to generate a bounded jerk trajectory with the minimum running time as the objective. The search performance of the WOA under complex constraints, including the search capability of trajectory planning symmetry, was improved by the following strategies: a dual-population search, including a new communication mechanism to prevent falling into the local optimum; a mutation centroid opposition-based learning, to improve the diversity of the population; and an adaptive inertia weight, to balance exploration and exploitation. Simulation analysis showed that the MWOA generated a trajectory with a shorter run-time and better symmetry and robustness than the WOA. Finally, a pilot rehabilitation session on a healthy volunteer using an upper extremity exoskeleton rehabilitation robot was completed safely and smoothly along the trajectory planned by the MWOA. The proposed algorithm thus provides a feasible scheme for isokinetic rehabilitation trajectory planning of upper extremity exoskeleton rehabilitation robots.

The optimization of clutch engagement strategies is of great significance for improving vehicle power performance, fuel economy, and driving comfort. Traditional control strategies are difficult to adapt to complex working conditions and lack coordinated optimization of fuel and clutch. This paper proposes a multi-objective optimization method for clutch engagement strategies based on the Deep Deterministic Policy Gradient (DDPG) algorithm. A simulation environment is constructed, which includes a vehicle longitudinal dynamics model, clutch state switching logic, and a reinforcement learning agent. A multi-dimensional state space and action space are designed, and a composite reward function combining power performance, fuel economy, and comfort is developed to achieve multi-objective optimization of the fuel–clutch coordination curve. Experimental results show that the optimized engagement strategy significantly reduces sliding friction power (by 94.07%), power interruption speed (by 8.75%), and jerk (with a maximum reduction of 35.6%), while the average fuel consumption per distance is reduced by 0.39%. Through weight sensitivity analysis, it is found that when the weight of fuel economy is 0.3 and the weight of power performance is 0.5 (Scheme P5E3), the optimal balance among multiple objectives can be achieved. This study provides a new theoretical framework and engineering practice reference for the intelligent control of clutches.

This study proposes a capacitance-based fatigue driving recognition method. The proposed method encompasses four principal phases: signal acquisition, pre-processing, blink detection, and fatigue driving recognition. A measurement circuit based on the FDC2214 is designed for the purpose of signal acquisition. The acquired signal is initially subjected to pre-processing, whereby noise waves are filtered out. Subsequently, the blink detection algorithm is employed to recognize the characteristics of human blinks. The characteristics of human blink include eye closing time, eye opening time, and idle time. Lastly, the BP neural network is employed to calculate the fatigue driving scale in the fatigue driving recognition stage. Experiments under various working and light conditions are conducted to verify the effectiveness of the proposed method. The results show that high fatigue driving recognition accuracy (92%) can be obtained by the proposed method under various light conditions.