Printed circuit board (PCB) assembly, on the basis of surface mount electronic component welding, is one of the most important electronic assembly processes, and its defect detection is also an important part of industrial generation. The traditional two-stage target detection algorithm model has a large number of parameters and the runtime is too long. The single-stage target detection algorithm has a faster running time, but the detection accuracy needs to be improved. To solve this problem, we innovated and modified the YOLO11n model. Firstly, we used the Retention Block (RetBlock) to improve the C3K2 module in the backbone, creating the RetC3K2 module, which makes up for the limitation of the original module’s limited, purely convolutional local receptive field. Secondly, the neck structure of the original model network is fused with a Multi-Branch Auxiliary Feature Pyramid Network (MAFPN) structure and turned into a multi-branch auxiliary neck network, which enhances the model’s ability to fuse multiple scaled characteristics and conveys diverse information about the gradient for the output layer. The improved YOLO11n model improves its mAP50 by 0.023 (2.5%) and mAP75 by 0.026 (2.8%) in comparison with the primitive model network, and detection precision is significantly improved, proving the superiority of our proposed approach.

Summary: This paper presents a new approach to synthesize multi-loop mechanisms with three translational displacement parameters based on virtual-loop theory and Assur groups. The approach used kinematic links as a generalized link group added one-by-one to the output link, which further extends the unified link groups in the plane and space. Firstly, the concept of infinitesimal displacement parameters is introduced to describe the displacement parameters. The dependence on the change in the degree of freedom (DOF) and displacement parameters of the output link after adding a 0-DOF generalized link group is established. Then, the link groups with three displacement parameters are synthesized, and the intersection operation rules are given. The single-loop mechanism is synthesized under two circumstances. The 1-, 2-, and 3-DOF dual-loop mechanisms are obtained by adding corresponding generalized link groups. Finally, the multi-loop mechanisms are obtained by adding corresponding generalized link groups. Some novel mechanisms are synthesized to illustrate the effectiveness of the proposed approach.

This article proposes a family of spatial three translational and one rotational parallel mechanisms (PMs) for pick-and-place operation. Their features are one independent rotation of the mechanism with four identical limbs, which are provided by the four revolute joints on the moving platform. The rotational capability of the PMs has a range of at least 180°. This article focuses on the synthesis of the PMs and kinematics analysis of the 4-P(2-SS)R parallel mechanism. First, based on the Lie group theory, three parallelograms are used in designing the PMs. The limbs are listed and two types of three translational and one rotational PMs are synthesized. Then, a typical 4-P(2-SS)R PM is selected, the 6 × 6 Jacobian matrix and the 6 × 6 × 6 Hessian matrix of the mechanism are derived for solving the displacement, velocity, and acceleration of the mechanism. Finally, singularity configurations are disclosed from the 6 × 6 Jacobian matrix, and the workspace of the mechanism is provided to illustrate the high rotational capability.

The article proposes a general approach for solving the unified formulas of the constrained Jacobian matrix and constrained Hessian matrix of three translational and one rotational (3T1R) parallel manipulators (PMs). First, based on the constraint properties of PMs, the constrained couples of the limbs are divided into two categories: invariable and variable constrained couples. The unified 2 × 6 constrained Jacobian matrix and 2 × 6 × 6 constrained Hessian matrix of the 3T1R PM are established. Then, a typical 2-(UPU)²R PM is selected, and the 6 × 6 Jacobian matrix and 6 × 6 × 6 Hessian matrix of the mechanism are deduced. The singularity of the mechanism is analyzed based on the 6 × 6 Jacobian matrix. Finally, the kinematics are derived, and results demonstrating the proposed method were provided.

Robotic exoskeletons with different actuators have been extensively investigated for decades, but designing a robot joint with compact structure, sufficient output torque, and satisfactory kinematic performance still remains a challenge. The paper introduces the mechanical design and control scheme of a novel joint with a circular rotary electro-hydraulic actuator in order to provide walking assistance for the hip joint. First, measurement and analysis techniques are applied in the gait recognition to determine an accurate swing range of the target joint and provide essential information for the structure design of the hydraulic rotary drive joint. Secondly, the functional structure and working principle of the hydraulic rotary joint are described in detail and the corresponding mathematical model is established for the valve-controlled hydraulic servo system. In addition, an adaptive sliding mode control (ASMC) method is proposed for position-tracking and improving the anti-disturbance ability of the controlled system. The global asymptotic stability and finite-time convergence of the closed-loop system are proved by Lyapunov stability theory. Finally, the feasibility and position-tracking performance of the designed ASMC controller were experimentally verified. Compared to the PID controller and the conventional sliding mode control scheme, the proposed controller improved the position-tracking control accuracy by 20%.