A nine-grid large model-based natural language interactive navigation and obstacle avoidance and grabbing method for a robot dog

By adopting a hierarchical modular system architecture and ROS2 communication, combined with natural language understanding, SLAM mapping and YOLOv8 visual inspection, the natural language interactive navigation and grasping of the quadruped robot dog was realized. This solved the problem of traditional quadruped robot dog control relying on professional operation and independent module design, and improved the efficiency and accuracy of navigation and grasping.

CN122281907APending Publication Date: 2026-06-26伊佳阳
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610396631.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-30
Publication Date
2026-06-26

Smart Images

  • Figure CN122281907A_ABST
    Figure CN122281907A_ABST
Patent Text Reader

Abstract

This invention discloses a natural language interactive navigation, obstacle avoidance, and grasping method for a robotic dog based on a nine-grid large model. The method includes the following steps: Step S1: Parsing user commands to extract key information and publishing standardized ROS2 topics; Step S2: Processing radar point clouds and fusing multi-sensor data to establish a coordinate transformation chain; Step S3: Navigating and avoiding obstacles to the target location based on the SLAM and Nav2 framework; Step S4: Completing target recognition and precise 3D positioning using a wrist camera and YOLOv8; Step S5: Generating a grasping action sequence and achieving closed-loop control of the robotic arm through smooth trajectory planning. This invention integrates nine-grid large model semantic parsing and a hierarchical modular architecture with multiple core technologies to achieve natural language interactive driving for the robotic dog, as well as the coordinated completion of navigation, obstacle avoidance, and object grasping, significantly reducing the operational threshold and improving grasping accuracy and motion safety.
Need to check novelty before this filing date? Find Prior Art