Cooperative grasping feature-based grasping control method and device, equipment and medium
By acquiring tactile and mechanical data of the target object, collaborative grasping features and feature weights are generated, a grasping model is constructed, and grasping force and posture are dynamically adjusted in combination with real-time feedback data. This solves the problem of independent tactile and mechanical perception during robot grasping, and achieves precise grasping of complex objects and improved safety.
Patent Information
- Application Number
- CN202511112514.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-08
- Publication Date
- 2026-09-11
- Estimated Expiration
- 2045-08-08
AI Technical Summary
In existing technologies, tactile and mechanical perception are independent during robot grasping, lacking a collaborative processing mechanism. This results in an inability to dynamically adjust grasping force and posture, an inability to adapt to the complex and diverse physical characteristics of objects, and a risk of grasping failure or object damage.
By acquiring the surface tactile data, interactive mechanical data, and physical parameters of the target object, collaborative grasping features and feature weight vectors are generated, a grasping model is constructed, and a grasping control sequence and collaborative adaptation coefficients are generated through constraint optimization by combining the initial grasping strategy and real-time tactile feedback data, and the grasping force and posture are dynamically adjusted.
It enables precise grasping and control of different objects, solves the problem of inflexible grasping response, improves dynamic adaptation and fine adjustment capabilities, and reduces the risk of damage.