Robot-Mounted Camera Position Detection During Motion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robot systems with cameras attached for object position detection face inefficiencies due to the need for stationary camera positions, leading to increased time delays and reduced accuracy when dealing with multiple overlapping objects.

Innovation Solution

The method involves creating images while the camera is moving, with real-time data from the robot controller providing camera position and orientation, allowing for faster and more accurate object position determination using image data from multiple perspectives and pattern recognition algorithms, and enabling direct feedback for camera repositioning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the robot stops at different positions to create images from multiple perspectives, then the position detection accuracy is improved, but the detection time is significantly increased

Engineering Contradiction:
Improveposition detection accuracyVSAvoiddetection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent transitions from a static imaging approach (robot stopping at predetermined positions) to a dynamic imaging approach (capturing images during continuous robot movement). The camera captures images while the robot moves through the workspace, eliminating stopping time while still acquiring images from multiple perspectives. This dynamic capture method maintains measurement precision by obtaining images from various positions without incurring the time penalty of repeated stops.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements continuous image capture during robot movement rather than intermittent capture at stopped positions. The useful action of image acquisition continues uninterrupted as the robot moves, ensuring that images from multiple perspectives are obtained without breaking the motion sequence. This continuity eliminates the time loss associated with stopping and starting the robot multiple times while maintaining the ability to detect object positions accurately from multiple viewpoints.

Inventive Principle:
Principle #20Continuity of useful action

2Productivity

If images are captured during continuous robot movement, then the detection time is reduced, but the complexity of coordinate transformation and position calculation increases

Engineering Contradiction:
Improvedetection speedVSAvoidcoordinate transformation complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent incorporates real-time feedback from the robot controller regarding the camera's position and orientation during movement. This feedback mechanism provides continuous data about the camera's spatial state, which is essential for accurate coordinate transformations. By utilizing this feedback, the system can dynamically adjust calculations to account for the changing camera perspective, managing the complexity through systematic use of available positional data rather than requiring overly complex pre-planned transformation models.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent handles coordinate transformation complexity by dynamically adjusting transformation parameters based on the camera's real-time position and orientation. Rather than using fixed transformation matrices, the system modifies transformation parameters continuously as the camera moves, allowing accurate position calculation from moving perspectives. This approach manages complexity by using adaptable parameters that reflect the actual camera state, simplifying the overall calculation process compared to rigid pre-programmed transformations.

Inventive Principle:
Principle #35Parameter changes

3Device complexity

If a single camera is mounted on the robot, then the system complexity is reduced compared to multiple fixed cameras, but the ability to capture objects from all perspectives is limited

Engineering Contradiction:
Improvesystem complexityVSAvoidperspective coverage
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent makes a single camera universal by utilizing it in multiple positions and orientations through robot movement. Rather than requiring multiple fixed cameras each dedicated to a specific viewpoint, the single camera attached to the robot can be positioned anywhere in the workspace and oriented in any direction needed. This multi-functional use of one camera achieves the same versatility as multiple cameras while significantly reducing system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent transforms a static single-camera system into a dynamic one where the camera's position and orientation change continuously with robot movement. This dynamic positioning allows the single camera to access multiple perspectives that would otherwise require multiple fixed cameras. The camera adapts its viewpoint dynamically based on the robot's motion, achieving comprehensive perspective coverage with minimal hardware.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2229262B2Determining the position of an object
Publication Date: 2015.02.11 ABB AG(DE)
  • EP2229262B2 patent drawingFigure 1
  • EP2229262B2 patent drawingFigure 2

AI summary

The invention relates to a method for determining the position of at least one object (22) present within a working range of a robot (12) by an evaluation system (16), wherein an image of at least one part of the working range of the robot (12) is generated by a camera (18) mounted on a robot (12). The image is generated during a motion of the camera (18) and image data are fed to the evaluation system (16) in real time, together with further data, from which the position and/or orientation of the camera (18) when generating the image can be derived. Said data are used for determining the position.