Robot Arm Pose Error Detection Using Visual Marker Association
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robot systems for remote operations face challenges in accurately detecting pose errors of operating arms in real-time, which affects the precision and reliability of motion control and human-computer interaction.
Innovation Solution
An error detection method that involves obtaining a target pose of the operating arm's end, acquiring positioning images, recognizing pose and angle identifications, determining the actual pose, and generating control signals based on error conditions, utilizing a computer device with processors and sensors to facilitate real-time error detection and control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If motion conversion control is used between master manipulator and operating arm, then remote operation capability is achieved, but pose error detection accuracy deteriorates
Solution Approach 1:
The patent introduces pose identifications (visual markers) as intermediary objects attached to the operating arm. These markers serve as mediators between the operating arm and the detection system, enabling accurate pose measurement through image recognition without directly measuring the arm's position and orientation. The marker's known geometric features provide reference points for calculating the operating arm's pose with high precision.
Solution Approach 2:
The patent creates a visual copy of the operating arm's pose information through pose identifications that can be captured by cameras. Instead of directly measuring the physical arm, the system captures an optical copy (image) of the marker attached to the arm, processes this visual information, and derives the pose data from the image, thereby achieving accurate remote pose detection.
2Manufacturing precision
If real-time pose error detection is implemented, then operation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent replaces complex mechanical measurement systems with an optical vision-based detection system. Instead of using encoders, sensors, or mechanical feedback devices on the operating arm, the system uses cameras to capture images of pose identifications and computationally determines pose information, significantly reducing mechanical complexity while maintaining real-time detection capability.
Solution Approach 2:
The system uses visual copying through image capture to obtain pose information. By attaching markers that can be recognized in images and processing these visual copies, the system achieves real-time pose detection without requiring complex physical sensing infrastructure, thereby improving operation accuracy while controlling system complexity.
3Measurement precision
If pose identifications are recognized in positioning images, then actual pose determination is achieved, but error detection time increases
Solution Approach 1:
The patent applies preliminary action by pre-defining the geometric features and positions of pose identifications on the operating arm before operation. The marker's structure, size, and marker-point arrangements are predetermined and known to the system. This allows the recognition algorithm to quickly match observed features against pre-stored templates, significantly reducing computation time while maintaining accurate pose determination.
Solution Approach 2:
The patent uses distinct visual features (color patterns, geometric shapes) of pose identifications to enable rapid differentiation and recognition. By designing markers with high-contrast, easily distinguishable visual characteristics, the system can quickly identify and process pose information from images, reducing detection time while ensuring accurate pose measurement through feature recognition.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present disclosure relates to the field of error detection technology, disclosing an error detection method and a robot system. The error detection method includes: obtaining a target pose of an end of an operating arm; acquiring a positioning image; recognizing, in the positioning image, a plurality of pose identifications located on the end of the operating arm; recognizing, based on the plurality of pose identifications, an angle identification located on the end of the operating arm, the angle identification having a position association relationship with a first pose identification of the plurality of pose identifications; determining, based on the angle identification and the plurality of pose identifications, an actual pose of the end of the operating arm; and generating, in response to the target pose and the actual pose meeting an error detection condition, a control signal related to a fault.