AR Robot Teaching With RGB-D Path Planning and Collision Checking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robot programming methods, such as online teaching, offline teaching, and autonomous programming, face challenges including low accuracy, complex processes, high costs, and poor adaptability to dynamic environments, especially in personalized product customization and mixed model production.
Innovation Solution
A system and method for robot teaching using RGB-D images and a teach pendant, which eliminates the need for complex position and posture tracking systems, employing AR simulation to generate robot programs, and includes a virtual-real collision detection module to prevent physical collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If online teaching programming is used to manually drag robot end effector or operate teach pendant, then the robot can be programmed to follow a target path, but the programming accuracy is hard to be ensured and the process is complicated with low efficiency
Solution Approach 1:
The patent uses AR technology to create a virtual copy of the robot and working environment, allowing programmers to teach and simulate robot paths in a virtual model that mirrors the physical system. This virtual copying enables precise path planning without physical trial-and-error, improving both accuracy and efficiency
Solution Approach 2:
The patent introduces an AR-based virtual reality system as an intermediary between the programmer and the physical robot. This intermediary layer allows for intuitive visual programming through virtual scene manipulation, eliminating the complexity of traditional teach pendant operations while maintaining high programming accuracy
2Productivity
If offline teaching programming is used to establish physical robot model and 3D model in virtual environment, then the virtual robot can be driven to simulate working process, but it requires professional technicians to build virtual model with large modeling workload and difficult operation
Solution Approach 1:
The patent enables the system to automatically capture and process real-world robot and environment data to generate virtual models automatically. This self-service approach eliminates the need for manual modeling by professional technicians, reducing both workload and complexity while maintaining model accuracy
Solution Approach 2:
The patent replaces the manual mechanical process of building 3D models with automated digital capture and processing systems. Using cameras and computational algorithms, the system automatically generates accurate virtual models from real-world images, substituting labor-intensive manual modeling with automated computational processes
3Extent of automation
If autonomous programming is used to detect physical working environment using visual sensors, then the robot can independently complete tasks without external control, but it has poor environment adaptability and high visual servoing cost
Solution Approach 1:
The patent performs preliminary environment modeling and path planning in the virtual AR scene before executing tasks in the physical environment. By pre-processing and pre-planning in the virtual space, the system improves adaptability to different environments without requiring complex real-time visual servoing, reducing costs while maintaining automation
Data Source
AI summary
A system for robot teaching based on RGB-D images and a teach pendant, including an RGB-D camera, a host computer, a posture teach pendant, and an AR teaching system which includes an AR registration card, an AR module, a virtual robot model, a path planning unit and a posture teaching unit. The RGB-D camera collects RGB images and depth images of a physical working environment in real time. In the path planning unit, path points of a robot end effector are selected, and a 3D coordinates of the path points in the basic coordinate system of the virtual robot model are calculated; the posture teaching unit records the received posture data as the postures of a path point where the virtual robot model is located, so that the virtual robot model is driven to move according to the postures and positions of the path points, thereby completing the robot teaching.