Most industrial applications concern output torque, rotor mass, and utilization of rare Earth permanent-magnet (PM) material of electrical machines. Generally, it is challenging to improve these aspects simultaneously. The objective of this article is to propose a novel concentrated-flux-type PM machine with irregular magnets and iron poles (CPMIP). The configuration has following advantages. First, the employment of quasitrapezoid magnets with opposite magnetization direction and large cross-section helps to enhance the flux-focusing effect and thus the torque generation. Second, the quasitriangular iron poles can improve the magnetic self-shielding effect well, and thus the rotor yoke can be removed completely and the rotor mass could be reduced significantly. Third, due to the reduction of PM volume and the improvement of torque output, the PM utilization ratio of electrical machine can be increased greatly. The schematic structure and pole patterns of the proposed electric machine are presented. Subsequently, the equivalent magnet circuit is established to validate the performance improvement of the CPMIP machine. Then, the design improvement with offset iron-pole arc is proposed to suppress torque ripple and balance torque output and PM utilization ratio capability well. Following that, an electrical machine with the irregular magnets and iron poles is designed. Then, the electromagnetic characteristics of the proposed pole designs are compared with those of the electric machines with conventional radial, Halbach, and consequent-pole patterns. Finally, the research prototype is developed, and experiments are conducted to verify the design concept and analysis of CPMIP machine.

When a mobile robot is required to perform tasks in the unknown and complex environment, it is critical to have the ability of dynamic obstacle avoidance. However, conventional deep deterministic policy gradient (DDPG) for collision-free navigation can only perceive a fixed number of dynamic obstacles, and thus it cannot adapt to the stochastic working scenario. To overcome the limitation, an improved DDPG algorithm is proposed in this study. It is an exploration to implement the DDPG with long short-term memory (LSTM) network-based encoder to achieve dynamic obstacle avoidance for the mobile robot in the stochastic working scenario, which can encode the variable number of obstacles into a fixed-length representation. Specifically, to facilitate the LSTM network-based encoder, one safe processing rule is designed to guarantee the entire information of the observable obstacles can be represented completely. The LSTM network-based encoder takes the latest environment information of observable obstacles by employing the safe processing rule and generates the fixed length state vector. In addition, continuous state space for mobile robot and obstacles, as well as reward function and action space are designed. Both simulations and experiments are conducted and the results verify that the improved DDPG algorithm can achieve collision-free trajectory with multiple dynamic obstacles well. It helps to reduce the path distance and motion time effectively.

This article presents a novel radial-flux surface-mounted PM (SPM) machine with quasi-cylindrical pole pattern (QCPP). With the proposed configuration, the cogging torque and torque ripple of surface-mounted machines could be greatly decreased due to the adoption of large eccentric distance and the reduction of high-order airgap flux density harmonics. Moreover, the increase of fundamental flux component helps to improve torque generation well. Additionally, the magnet poles are semi-embedded into the rotor, and thus the sleeve in conventional SPM machines is no longer necessary, which helps to reduce the airgap size and improve the torque output further. Besides, the quasi-cylindrical magnet could offer lower rotor eddy loss and higher anti-demagnetization capability than conventional radial and eccentric pole patterns. Studies have been conducted on the novel structure to validate its advantages. Firstly, the topological structure and working principle of the QCPP machine are presented. Then, the influence of the QCPP structural parameters on the torque performance is investigated. Next, an electrical machine with quasi-cylindrical pole pattern is designed. Subsequently, the electromagnetic characteristics, such as flux density distribution, back EMF, average torque, cogging torque, torque ripple, eddy loss in sleeve, rotor stress and anti-demagnetization capability, of the proposed machine are analyzed and compared with those of conventional SPM machine with eccentric magnetic poles and skewing slots. Finally, one research prototype is developed, and experiments are carried out to evaluate the design concept and performance of the proposed QCPP machine.

A linear oscillating motor provides short-stroke reciprocating motion directly and is widely used in linear actuation systems. However, with the increase of actuation frequency to achieve high power density, the large mover mass becomes the main obstacle. Existing solutions either sacrifice force density or increase fabrication difficulties for high dynamic performance. In this paper, a novel compound Halbach magnet array is proposed for moving-magnet tubular linear oscillating motor in high-frequency applications. In this structure, dual-layer magnets are utilized for better self-shielding effects. Thus, the conventional back-iron in the mover is no longer needed, which contributes to both high force density and low mover mass simultaneously. The analytical model is first set up based on harmonic function method. Then, finite-element-method results are used for validation and comparison with conventional designs. Based on the verified model, electromagnetic structure optimization is conducted, which leads to the prototype structure. Finally, experimental results on electromagnetic performance are provided for validation of the proposed motor with a novel magnet array.

Human–robot-collaboration requires robot to proactively and intelligently recognize the intention of human operator. Despite deep learning approaches have achieved certain results in performing feature learning and long-term temporal dependencies modeling, the motion prediction is still not desirable enough, which unavoidably compromises the accomplishment of tasks. Therefore, a hybrid recurrent neural network architecture is proposed for intention recognition to conduct the assembly tasks cooperatively. Specifically, the improved LSTM (ILSTM) and improved Bi-LSTM (IBi-LSTM) networks are first explored with state activation function and gate activation function to improve the network performance. The employment of the IBi-LSTM unit in the first layers of the hybrid architecture helps to learn the features effectively and fully from complex sequential data, and the LSTM-based cell in the last layer contributes to capturing the forward dependency. This hybrid network architecture can improve the prediction performance of intention recognition effectively. One experimental platform with the UR5 collaborative robot and human motion capture device is set up to test the performance of the proposed method. One filter, that is, the quartile-based amplitude limiting algorithm in sliding window, is designed to deal with the abnormal data of the spatiotemporal data, and thus, to improve the accuracy of network training and testing. The experimental results show that the hybrid network can predict the motion of human operator more precisely in collaborative workspace, compared with some representative deep learning methods.

Torque motor is one key component that directly influences the dynamic performance of jet pipe servo valve in aircraft. In this paper, a novel torque motor with hybrid-magnetization pole arrays is proposed. By changing the magnetization patterns of permanent magnets, the torque motor can significantly improve the output torque by range of 47% - 52% compared with traditional designs, while maintaining the system size and mass. The design concept and operating principle of the torque motor is presented. The magnetic field distribution is formulated analytically with equivalent magnetic circuit. Different from conventional study, the flux leakage of the permanent magnets and coils is included to improve the model precision. Subsequently, the output torque is derived mathematically from the air gap flux. Following that, the numerical calculation is conducted to validate the mathematical models of magnetic field and torque output. The design optimization is then carried out. One research prototype that can be mounted with either conventional magnet or the proposed hybrid array has been developed. The test rigs are constructed and experiments are conducted on the prototype. Both numerical computation and experimental results verify the significant improvement of torque generation of the proposed hybrid magnetization torque motor.