A predictive energy management strategy considering travel route information is proposed to explore the energy-saving potential of plug-in hybrid electric vehicles. Extreme Learning Machine is used as a short-term speed predictor, and the battery temperature is added as an optimization term to the cost function. By comparing the training datasets, it is found that using the real-world historical speed information for training can achieve higher prediction accuracy than using typical standard driving cycles. The speed predictor trained based on the data considering travel route information can further improve the prediction accuracy. The impact of battery temperature on the total cost is also analyzed. By adjusting the temperature weighting coefficient of the battery, a balance between economy and battery aging can be achieved. In addition, it is found that the ambient temperature also affects vehicular energy consumption. Finally, the proposed method is compared to PMP, MPC, and CD-CS methods, showing its effectiveness and practicability.

Committed to optimizing the fuel economy of hybrid electric vehicles (HEVs), improving the working conditions of the engine, and promoting research on deep reinforcement learning (DRL) in the field of energy management strategies (EMSs), this article first proposed a DRL-based EMS combined with a rule-based engine start–stop strategy. Moreover, considering that both the engine and the transmission are controlled components, this article developed a novel double DRL (DDRL)-based EMS, which uses a deep Q-network (DQN) to learning the gear-shifting strategy and uses a deep deterministic policy gradient (DDPG) to control the engine throttle opening, and the DDRL-based EMS realizes the multiobjective synchronization control by different types of learning algorithms. After off-line training, the simulation result of the online test shows that the fuel consumption gaps of the proposed DRL- and DDRL-based EMSs are −0.55% and 2.33% compared to that of the deterministic dynamic programming (DDP)-based EMS by overcoming some inherent flaws of DDP, respectively. The computational efficiency has been significantly improved, and the average output time per action is 0.91 ms. Therefore, the control strategy that combines learning- and rule-based controls and the multiobjective control strategies both have the potential to ensure optimization and real-time efficiency.

Magnetic soft robots have shown great potential for biomedical applications due to their high shape reconfigurability, motion agility, and multi-functionality in physiological environments. Magnetic soft robots with multi-layer structures can enhance the loading capacity and function complexity for targeted delivery. However, the interactions between soft entities have yet to be fully investigated, and thus the assembly of magnetic soft robots with on-demand motion modes from multiple film-like layers is still challenging. Herein, we model and tailor the magnetic interaction between soft film-like layers with distinct in-plane structures, and then realize multi-layer soft robots that are capable of performing agile motions and targeted adhesion. Each layer of the robot consists of a soft magnetic substrate and an adhesive film. The mechanical properties and adhesion performance of the adhesive films are systematically characterized. The robot is capable of performing two locomotion modes, i.e., translational motion and tumbling motion, and also the on-demand separation with one side layer adhered to tissues. Simulation results are presented, which have a good qualitative agreement with the experimental results. The feasibility of using the robot to perform multi-target adhesion in a stomach is validated in both ex-vivo and in-vivo experiments.

Jumping is an important locomotion function to extend navigation range, overcome obstacles, and adapt to unstructured environments. In that sense, continuous jumping and direction adjustability can be essential properties for terrestrial robots with multimodal locomotion. However, only few soft jumping robots can achieve rapid continuous jumping and controlled turning locomotion for obstacle crossing. Here, we present an electrohydrostatically driven tethered legless soft jumping robot capable of rapid, continuous, and steered jumping based on a soft electrohydrostatic bending actuator. This 1.1 g and 6.5 cm tethered soft jumping robot is able to achieve a jumping height of 7.68 body heights and a continuous forward jumping speed of 6.01 body lengths per second. Combining two actuator units, it can achieve rapid turning with a speed of 138.4° per second. The robots are also demonstrated to be capable of skipping across a multitude of obstacles. This work provides a foundation for the application of electrohydrostatic actuation in soft robots for agile and fast multimodal locomotion.

This paper presents a reliable intelligent path following control method for a robotic airship subject to sensor faults. First, an adaptive backstepping sliding mode controller is designed based on the six degrees of freedom model of airship, where the backstepping technique is used to obtain the desired velocities, and the adaptive sliding mode method is adopted to deal with the unknown model uncertainties. The stabilization of the whole path following control system is studied based on Lyapunov stability theory. Specially, since the measured data in the airship control system are interrelated, a data-driven soft sensor based on adaptive neuro-fuzzy inference system is designed to detect and isolate the sensor fault. Based on the developed detection and isolation mechanism, the airship control system still work well in the presence of sensor fault. Finally, simulations are given to demonstrate the effectiveness of the proposed path following strategy.