Since the trash can is not sanitary and fails to do the garbage classification, a household intelligent trash can based on SCM is designed. Firstly, the hardware of the intelligent trash can is designed, whose functio...
详细信息
This research includes the creation and execution of a Self-Balancing Robot (SBR) that is operated by stepper motors. The robot39;s primary operation depends on the accurate manipulation of pulse-driven stepper moto...
详细信息
Graph Neural Network Reinforcement Learning (GNN RL) has been widely used in robotic control applications. In designing GNN RL for robots, while kinematic topology has been traditionally used as the message-passing to...
详细信息
ISBN:
(纸本)9798350307573;9798350307566
Graph Neural Network Reinforcement Learning (GNN RL) has been widely used in robotic control applications. In designing GNN RL for robots, while kinematic topology has been traditionally used as the message-passing topology, the effect of different structural designs of message-passing topology on acquired actions has not well been studied. In this paper, we evaluated those effects on a screw task by a D'Claw ROBEL hand using three types of message-passing topologies: series, parallel and series+parallel topology. As a result, we have found that parallel topology promotes the acquisition of more similar actions in a cooperative way due to homogeneous message passings to each finger, and also series topology promotes more diverse actions in a separate way due to heterogeneous message passings to each finger. Moreover, series+parallel topology acquire diverse and cooperative actions simultaneously because it generates both homegeneous and heterogeneous messages for each finger.. In future work, we plan to apply this technique to multi-task RL problems.
In order to solve the problems of classical convolutional neural networks in coal and gangue recognition, such as large calculation amounts, slow recognition speeds, and low recognition accuracy, a method for coal and...
详细信息
Series-type and ball-and-socket are currently two typical architectures of hip joint actuators. However, they either rely on advancements and applications in materials science or compromise on structural design, faili...
详细信息
ISBN:
(纸本)9798350307573;9798350307566
Series-type and ball-and-socket are currently two typical architectures of hip joint actuators. However, they either rely on advancements and applications in materials science or compromise on structural design, failing to accurately mimic the motion of the human hip joint. Therefore, we designed a three-degree-of-freedom (3-DOF) dexterous joint actuator based on multi-drive hierarchical output and spherical rotation mechanism for better kinematics performance, high power density, and high load-withstand capability. Three identical motors were installed in parallel on the motor mounting frame, with two motors coupled and controlled a pair of differential bevel gears, while another motor transmitted torque through two spherical gears, coupled with a secondary frame mounted on a large bevel gear, ultimately achieving spherical rotation with a fixed point as the center. 3D simulations demonstrated that this new design could achieve a 120-degree rotation range across three degrees of freedom, with high load-withstand capability, successfully meeting project expectations and demonstrating its usability. This design closely approximated the overall performance of the human hip joint in terms of driving performance, motion characteristics, and control logic. It provides theoretical support for addressing existing engineering bottlenecks in joint actuators and has promising potential for various robotic applications.
Current synchronous control techniques for non-linear feedback in permanent magnet synchronous motors have large synchronous control time delays due to weak sensor signals. For this reason, a reinforcement learning ba...
详细信息
Reinforcement learning has attracted much attention in the field of mechanical control, but it has rarely been applied in projects requiring high precision and short control cycle. In order to realize the high frequen...
详细信息
Piezoelectric cantilever actuators have garnered significant attention in the Micro-electromechanical Systems (MEMS) domain due to their unmatched attributes, such as high resolution, speed, and accuracy. They have fu...
详细信息
In order to solve the internal coupling problem of the levitation module, it is necessary to design decoupling algorithms for the levitation system under the modular structure to suppress the coupling disturbance betw...
详细信息
An improved sliding mode control (ISMC) method based on linear extended state observer (LESO) is proposed to solve the trajectory tracking problem of underactuated ships with unknown states in the environment of distu...
详细信息
暂无评论