Camera Calibration Using Robot Trajectories for Subject Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing camera systems in warehouses and industrial facilities are not calibrated for advanced purposes like determining and tracking subject positions, requiring manual calibration which is inconvenient and expensive, and separate systems for surveillance and position tracking increase costs and failure risks.

Innovation Solution

A method for calibrating cameras using a mobile robot within a predefined coordinate system by minimizing reprojection errors, allowing the camera to be used for advanced purposes such as determining and tracking subject positions, and enabling automatic recalibration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual calibration is performed regularly, then camera accuracy for advanced purposes is improved, but convenience and cost increase

Engineering Contradiction:
Improvecamera accuracyVSAvoidconvenience
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The system performs self-calibration automatically by having the mobile robot execute predefined trajectories and capture images. The calibration process is autonomous, requiring no manual intervention, and uses the robot's own movement and imaging to optimize camera parameters and extrinsic calibration automatically.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The mobile robot executes predefined trajectories in advance to capture calibration images before actual operation. These preliminary images are used to optimize camera parameters and establish accurate calibration, ensuring the camera is ready for subsequent advanced applications without requiring manual calibration during operation.

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If separate systems are used for surveillance and position tracking, then functional requirements are met, but system complexity and cost increase

Engineering Contradiction:
Improvefunctional requirementsVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The camera system is designed to perform multiple functions: surveillance/monitoring and advanced position determination/tracking. By implementing automatic calibration using mobile robot trajectories, the same camera infrastructure supports both basic surveillance and advanced applications, eliminating the need for separate dedicated systems.

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

Solution Approach 2:

The patent combines surveillance camera functionality with advanced position tracking capabilities in a single integrated system. The calibration method enables the camera to serve dual purposes, merging what were previously separate systems into one unified infrastructure that handles both monitoring and precise position determination.

Inventive Principle:
Principle #5Merging (Combining)

3Ease of manufacture

If existing camera systems are used for advanced purposes, then cost is reduced, but calibration accuracy and reliability are insufficient

Engineering Contradiction:
ImprovecostVSAvoidcalibration accuracy
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The calibration system uses feedback from captured images of the mobile robot during trajectory execution to optimize camera parameters. The reprojection error between expected and actual image locations provides continuous feedback that drives the optimization process, ensuring high calibration accuracy and reliability through iterative improvement.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent replaces manual mechanical calibration procedures with an automated computational approach. Instead of physical manual adjustment, the system uses image processing, optimization algorithms, and the mobile robot's predefined trajectories to automatically calculate and optimize camera parameters, achieving both cost reduction and high reliability.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentEP4668213A1A method for calibrating a camera and methods for determining the location of a subject and methods for tracking a subject
Publication Date: 2025.12.24 VERITY AG
  • EP4668213A1 patent drawingFigure 1
  • EP4668213A1 patent drawingFigure 2
  • EP4668213A1 patent drawingFigure 3

AI summary

According to the present invention there is provided a method for calibrating a camera (102) which is in a position and orientation within a predefined coordinate system, the method comprising the steps of, (a) moving an object along a trajectory within the predefined coordinate system and within a field of view of the camera; (b) determining the physical location of the object (104), over time, within the predefined coordinate system (101); (c) capturing one or more images of the object (104) using the camera (102) as the object (104) is moving along the trajectory (105); (d) recording the time instant at which each of the respective one or more images are captured; (e) processing each of the one or more images to determine the location of the object (104) in the each of the one or more images, to provide an respective image location for each of respective image; (f) for each of the one or more images, determining a respective expected image location of the object in that image, wherein the expected image location is determined using an initial predefined estimate of camera parameters and the physical location of the object 104 in the predefined coordinate system (101) at the time instant corresponding to the time instant said respective image was captured; (g) optimizing estimates of camera parameters of the camera (102) and/or optimizing an estimate of the position of the camera (102) and/or optimizing an estimate of the orientation of the camera (102), by minimizing reprojection errors for each of said one or more images, wherein the reprojection error of a respective image is the difference between the expected image location of the object (104) for that image and the image location of the object (104) in said image. There is further provided methods for determining and methods for estimating, a physical location of a subject located within a predefined coordinate system, and methods for tracking the position of a subject, each of which use at least one camera which have been calibrated using the aforementioned method for calibrating a camera.