Multi-Objective Optimal Trajectory Planning for Robotic Arms Using Deep Reinforcement Learning.

This study investigated the trajectory-planning problem of a six-axis robotic arm based on deep reinforcement learning. Taking into account several characteristics of robot motion, a multi-objective optimization approach is proposed, which was based on the motivations of deep reinforcement learning and optimal planning. The optimal trajectory was considered with respect to multiple objectives, aiming to minimize factors such as accuracy, energy consumption, and smoothness. The multiple objectives were integrated into the reinforcement learning environment to achieve the desired trajectory. Based on forward and inverse kinematics, the joint angles and Cartesian coordinates were used as the input parameters, while the joint angle estimation served as the output. To enable the environment to rapidly find more-efficient solutions, the decaying episode mechanism was employed throughout the training process. The distribution of the trajectory points was improved in terms of uniformity and smoothness, which greatly contributed to the optimization of the robotic arm's trajectory. The proposed method demonstrated its effectiveness in comparison with the RRT algorithm, as evidenced by the simulations and physical experiments.

Research on Formation Conditions of the Ultrafine-Grained Structure of the Cylindrical Parts Manufactured by Power Spinning Based on Small Strains.

Two different methods, power spinning and annealing (PSA), quenching and power spinning followed by annealing (QPSA), for manufacturing the cylindrical parts with ultrafine-grained (UFG) structure were reviewed, the dislocation density and microstructural evolution during the two different processes of PSA and QPSA were further studied. The results show that the required strains for obtaining the UFG structure by power spinning is only 0.92 when the initial microstructure of the material is in the phase of lath martensite. The dislocation density and storage energy are increased to 10 times that of the blank after quenching and power spinning and decreased to the level of the blank after recrystallization annealing. Microstructures with fine grain size after quenching, storage energy of 1.8 × 105 kJ/m³ obtained after power spinning and second phase particle with nano-scale precipitated during annealing are the necessary formation conditions for manufacturing the cylindrical parts with UFG structure based on small strains. Compared with the original tubular blank, the mechanical properties of the spun parts with UFG structure improves significantly. The tensile strength and hardness of the spun parts manufactured by QPSA method is 815 MPa and 305 HV, respectively, and the elongation is 17.5%.