Reinforcement learning of shaft sleeve assembly strategy model and shaft sleeve assembly method and device
By using reinforcement learning to optimize the axle assembly process through axle assembly strategy model, combining safety and efficiency rewards, the problems of low assembly efficiency and insufficient safety are solved, achieving efficient and safe axle assembly.
CN117032079BActive Publication Date: 2026-06-16INST OF AUTOMATION CHINESE ACAD OF SCI
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INST OF AUTOMATION CHINESE ACAD OF SCI
- Filing Date
- 2023-08-16
- Publication Date
- 2026-06-16
AI Technical Summary
⚠Technical Problem
Existing technologies for assembling central shafts are inefficient and require a lot of manpower, making it difficult to ensure safety in complex and ever-changing assembly environments.
⚗Method used
A reinforcement learning approach is adopted for the bushing assembly strategy model. By acquiring the bushing assembly force and insertion depth, the assembly action amount is predicted by combining the initial model, and the model parameters are iterated based on safety rewards and efficiency rewards to optimize the assembly process.
🎯Benefits of technology
It improves the safety and efficiency of shaft assembly, can adapt to complex and ever-changing assembly environments, and reduces manual intervention.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117032079B_ABST
Abstract
The application provides a kind of bushing assembly strategy model reinforcement learning and bushing assembly method and device, wherein the method comprises: obtaining the assembly force of bushing in the previous step and the insertion depth of bushing;Based on the initial model, determine the predicted assembly action amount of the current step when the assembly force and insertion depth of the previous step are states;Based on the assembly action amount of the current step, update the assembly force and insertion depth of the current step, and determine the safety reward of the current step based on the assembly force of the current step, determine the efficiency reward of the current step based on the number of steps of the previous assembly round, the safety reward of the previous step and / or the assembly action amount of the current step;Based on the safety reward and the efficiency reward, the initial model is iterated to obtain the bushing assembly strategy model.The method and device provided by the application obtain a bushing assembly strategy model with good safety and high assembly efficiency, thereby simultaneously improving the safety and assembly efficiency of the bushing assembly based on the bushing assembly strategy model.
Need to check novelty before this filing date? Find Prior Art