Robot ordering method and system
By acquiring depth and color images, detecting target objects and cropping 3D point cloud data, using the pre-trained GraspNet network to predict candidate grasping poses, and combining speech recognition and visual language models for trajectory planning and motion execution, the problem of complex robot ordering methods and low control efficiency is solved, and efficient and safe robotic arm operation is achieved.
CN120645236APending Publication Date: 2025-09-16SHANGHAI MAJIKE IND INTELLIGENCE TECHNOLOGY CO LTD
0 Cites 4 Cited by
Patent Information
- Application Number
- CN202511161775.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-19
- Publication Date
- 2025-09-16
AI Technical Summary
Technical Problem
Existing robot ordering methods are complex and have low control efficiency, and there are problems with hardware driving and message transmission.
Method used
By acquiring depth and color images, detecting target objects and cropping 3D point cloud data, the pre-trained GraspNet network is used to predict candidate grasping poses. Trajectory planning and motion execution are combined with speech recognition and visual language models to achieve intelligent control of the robotic arm.
Benefits of technology
It improves the efficiency and stability of robot ordering, ensures the safety and continuity of robotic arm operations, and implements multi-level security protection throughout the entire process from prediction to execution.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention relates to the field of robots, and discloses a robot ordering method and system, and the method comprises the steps: obtaining a depth map and a color map of a current scene; obtaining a detection frame of a target object, and cutting out three-dimensional point cloud data of the corresponding object from the depth map according to a detection result; reasoning by taking the three-dimensional point cloud data as input of a reasoning model, and predicting candidate grabbing poses; and according to the predicted candidate grabbing poses, trajectory planning and motion execution are carried out. Reasonable action information can be generated according to information provided by the visual language model in combination with specific task requirements, and therefore control over the mechanical arm is achieved.
Need to check novelty before this filing date? Find Prior Art