Deep reinforcement learning (DRL) has been widely applied to robotic control, with trajectory tracking control being a particularly popular topic. However, existing research primarily focuses on simple models with low degrees of freedom (DOFs). The slidable-wheel omnidirectional mobile robot (SWOM), previously proposed by the authors, is an omnidirectional mobile robot characterized by passive moving components and 12 DOFs. The complexity of this structure presents significant challenges for controlling SWOM using DRL. In this paper, we propose a control framework that integrates DRL with hierarchical control to achieve DRL-based trajectory tracking for SWOM. The upper-level controller computes the required velocity of the robot body based on the error between the current and target positions, thereby regulating the wheel’s rotational speed. The lower-level controller consists of three DRL-trained sub-controllers, each responsible for steering one of SWOM’s three wheels, ensuring that the passive components of SWOM remain within an acceptable range. The design of the reward function in reinforcement learning, as well as the impact of parameter variations on training outcomes, is discussed. The effectiveness of the proposed controller is validated through numerical simulations conducted under two scenarios: one without considering wheel slippage and the other with wheel slippage taken into account. The proposed controller demonstrates satisfactory performance in both cases.

Fault diagnosis models based on data-driven method have always been a hot topic of research, promoting the intelligent development of rolling bearing fault diagnosis methods. However, data-driven fault diagnosis methods fail to directly account for fault mechanisms, thereby diminishing the credibility of their results. Therefore, a novel wavelet constrained physics-informed neural network (WCPINN) is proposed for bearing fault diagnosis. Firstly, an ensemble wavelet convolutional layer is proposed to extract physical information related to the health condition of bearings. Secondly, a novel physics-constrained loss is designed into the training stage to guide the parameters updating direction of the model. Two bearing datasets are utilized to validate the effectiveness of the proposed method. The results suggest that the is capable of fully focusing on the fault-related physical information within the original vibration signals, thereby endowing the model with a certain degree of feature credibility. The addition of physical constraints makes the training of the neural network more targeted, enhancing the stability and accuracy of the diagnosis results.

Aiming at the problems of low picking efficiency and fruit damage in existing apple-picking machinery, this paper presents an apple-picking robot equipped with a flexible clamping mechanism and a rigid cutting mechanism, designed specifically for standardized dwarf and densely planted apple orchards. The flexible clamping mechanism primarily consists of three fingers driven by the flexible spiral, each encased in a soft finger sleeve with a finned structure. Analyzing the simplified model of the flexible finger, the factors affecting its bending angle are identified. Based on these factors, the key structural parameters for achieving a single stable clamping of the fruit are determined. Tests on bending angle and output force confirm that the flexible finger meets the stability requirements for clamping apples. To verify the feasibility of trajectory planning and picking strategy, the kinematics equations of the robotic arm are established using the D-H method, the workspace is determined, and the trajectory of the robotic arm is simulated. To verify the performance of the picking robot, the picking test platform and the corresponding control system are built. The picking test results show that the robot achieved the average single-fruit picking time of 4.91 seconds, the picking success rate of 91.11%, and the picking damage rate of 4.89%. This study offers a reference for achieving nondestructive fruit picking using robots.

In machining processes using a 6-DOF robot, it is necessary to achieve continuous and smooth motion of the tool position and tool orientation along the target trajectory. When the tool orientation changes rapidly, joint angles may experience abrupt variations, leading to increased joint acceleration and jerk, structural vibrations, trajectory errors, and even abnormal operation stops. In the motion planning of a 6-DOF robot arm, the description of the tool orientation can be classified into cases where redundancy in the rotational degrees of freedom can be sufficiently ensured and cases where only limited redundancy is available. In the former case, the rotation around the tool axis can be freely adjusted, allowing conventional methods to smooth joint motion by utilizing this redundancy. However, in tasks where all three rotational degrees of freedom are constrained by task requirements, such redundancy cannot be ensured. To address this limitation, we propose a tool orientation planning method for continuous operations that minimize joint acceleration and joint jerk under limited redundancy conditions. The tool orientation is calculated through optimization using an objective function that evaluates motion smoothness, based on the target tool path and the task-specific rotational angle limits at each command point. The commanded tool tip position is kept fixed. Experimental validation was conducted using a 6-DOF industrial robot (LR Mate 200 iD/7L, FANUC), comparing commands generated from target surface normals and those generated by the proposed method. The results demonstrated the avoidance of abnormal stops, suppression of tool tip acceleration, and a reduction in joint jerk by up to approximately 70%, confirming the effectiveness of the proposed method.

In recent years, the demand for plate forging has been increasing with the demand for lighter weight automotive parts. The authors developed a method to reduce the load by applying lateral oscillation during compression, and demonstrated this load reduction during the forging of cylindrical specimens. This method was applied to the partial forging of a disc in the present study. Disc specimens with diameters of 8 and 14 mm were fabricated from 3 mm thick annealed pure aluminum plates, and the outer edge of each specimen was compressed to half its thickness. A linear actuator was used to apply lateral oscillation to the lower die during the second half of compression. For the disc specimens with a diameter of 8 mm, increasing the oscillation velocity under non-lubricated conditions resulted in a greater reduction in load. In the case using the 14-mm-diameter specimens, it was found that forging with grooved dies was an effective method to reduce the load, even under a lubricated condition. Finite element analyses revealed that when the compressed region was firmly gripped, lateral oscillation produced shear stress, which reduced the axial compressive stress. Experiments and finite element analyses using grooved dies were conducted on the 14-mm-diameter specimens by varying the lateral oscillation velocity. The relationship between the ratio of lateral oscillation velocity to axial compression velocity and the load-reduction ratio was investigated, and the experimental results showed good agreement with those obtained by finite element analysis. Measurements of the cross-sectional shape of the forged region confirmed that lateral oscillation caused local elongation along the die surface when the grooved dies were used.

The present study is part of a series aimed at enhancing the reliability of JIS B 1759, ”Estimation of tooth bending strength of cylindrical plastic gears.” This study examined the stress correction factor converting the nominal tooth bending stress into the local one. The current version has a problem with underestimating the load capacity of internal gears, but considering the contact ratio under load has alleviated the problem. Nevertheless, some issues remain, which led the present paper to determine an alternative stress correction factor through finite element analyses of internal virtual spur gears. Before discussing the alternative factor, examining the fillet profile of injection-molded internal spur gears led to proposing a single circular arc as the standard fillet. Based on that, FEAs of internal virtual spur gears with a single circular fillet determined the modified stress correction factor, which can relieve JIS B 1759 of the underestimations. The modified factor enables the standard to estimate load capacity regardless of external or internal gears, which improves the reliability of the standard.