Robot Trajectory Programming Using 3D Workspace Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for programming robots to perform trajectory points are often inaccurate and require a 3D CAD model or physical guidance, which can lead to collisions and increased costs due to the need for specialized knowledge and equipment.
Innovation Solution
A computer-implemented method using a sensor and display interface to acquire and manipulate 3D images of the environment, allowing for the identification and programming of robotic devices without relying on virtual models, by determining poses and trajectory points directly from real-world data, enabling intuitive and secure programming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional robot programming methods are used (physical guidance or 3D CAD models), then the robot can be programmed to perform trajectory points, but the process requires specialized knowledge, increases costs, and may lead to collisions due to inaccuracies
Solution Approach 1:
The patent uses a camera to capture images of the real environment and creates a visual copy of the workspace. Trajectory points are defined by selecting features in the captured image, which are then mapped to real-world coordinates. This eliminates the need for complex 3D CAD models while maintaining accuracy, as the system works directly with visual data from the actual environment.
Solution Approach 2:
The patent replaces physical guidance methods (mechanical teaching pendants, manual robot movement) with a vision-based system. The camera captures the environment, and trajectory points are defined through image processing and feature selection rather than physical robot movement. This substitution eliminates the need for specialized programming knowledge and reduces the risk of collisions during programming.
2Ease of manufacture
If 3D CAD models are used for programming, then trajectory points can be defined in virtual space, but costly software and models are required
Solution Approach 1:
The patent replaces expensive, permanent 3D CAD software with a simpler vision-based system using standard camera equipment. The system captures images of the real environment and processes them directly, eliminating the need for costly CAD software licenses and model creation. This makes robot programming accessible to smaller operations without specialized software investments.
Solution Approach 2:
The system automatically captures images of the workspace and enables trajectory point definition through direct image manipulation. The environment itself provides the reference data through the camera capture, eliminating the need for external CAD models or specialized programming tools. The system serves itself by using the actual workspace geometry as captured by the camera.
3Reliability
If physical guidance through trajectory points is used, then the robot learns the operating path, but collisions may occur during the teaching process
Solution Approach 1:
The patent captures images of the entire workspace beforehand and defines all trajectory points in the image space before executing the program. The camera captures the environment state, and all trajectory points are selected and defined in advance through image processing. This preliminary definition eliminates the need for gradual physical guidance, allowing the robot to execute the complete trajectory without human intervention and eliminating collision risks during programming.
Solution Approach 2:
The patent introduces a camera-based vision system as an intermediary between the programmer and the robot. Instead of directly guiding the robot through physical movement, the system captures images and defines trajectory points in the image space. This intermediary layer allows safe definition of all trajectory points before execution, eliminating the need for gradual teaching and preventing collisions during programming.
Data Source
AI summary
A method comprising identifying a robotic device and a calibration fixture in a vicinity of the robotic device; referencing the calibration fixture to a base of the robotic device to determine a first pose of the robotic device; receiving a 3D image of the environment, wherein the 3D image includes the calibration fixture; determining a second pose of the calibration fixture relative to the sensor; determining a third pose of the robotic device relative to the sensor based on the first pose and the second pose; receiving a plurality of trajectory points; determining a plurality of virtual trajectory points corresponding to the plurality of trajectory points based on the 3D image and the third pose; providing for display of the plurality of virtual trajectory points; and providing an interface for manipulating the virtual trajectory points.