Mobile Measurement Platform with Two-Stage Object Pose Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing coordinate measuring machines and robotic systems face challenges in achieving fast and precise positioning of measuring or machining tools relative to objects, particularly in industrial settings, due to their stationary nature and the need for the object to be brought to the machine, which limits throughput and accuracy.
Innovation Solution
A mobile platform with a kinematic system and an instrument head, equipped with sensors and a controller, uses a combination of initial and refined pose estimation methods, including machine learning and sensor fusion, to accurately position the instrument head on the target object, enabling precise alignment with an accuracy of up to ±0.1 mm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a stationary coordinate measuring machine is used, then measurement precision can be achieved, but the object must be brought to the machine and the setup is complex and time-consuming
Solution Approach 1:
Instead of bringing the object to a stationary measuring machine, the patent inverts the approach by making the measuring machine mobile and bringing it to the object. The mobile platform with mounted sensors and instruments moves to the workpiece, eliminating the need for complex object positioning and setup procedures while maintaining measurement capabilities.
Solution Approach 2:
The patent replaces traditional mechanical positioning systems with sensor-based detection and navigation. Instead of using complex mechanical guides and positioning mechanisms, the system uses sensors to detect the object's position and orientation, then navigates the mobile platform accordingly, reducing mechanical complexity.
2Productivity
If traditional robotic systems are used for positioning, then automation is achieved, but positioning speed and accuracy are insufficient for high-throughput industrial production
Solution Approach 1:
The patent implements a feedback mechanism where sensors continuously detect the object's position and orientation, and this information is fed back to the control system. The system compares the detected pose with the target pose and adjusts the mobile platform's position in real-time, enabling both high speed and high precision positioning through iterative correction.
Solution Approach 2:
The patent performs preliminary detection of the object's pose using sensors before final positioning. This preliminary action provides initial position and orientation information that guides the mobile platform's approach, allowing for faster and more accurate positioning by avoiding trial-and-error adjustments.
3Measurement precision
If multiple sensors are used for positioning, then accuracy improves, but the system complexity and computational requirements increase
Solution Approach 1:
The patent merges multiple sensor types (cameras, LIDAR, other sensors) into an integrated sensing system. Instead of treating each sensor independently, the system combines their data streams and processes them together to determine the object's pose, reducing overall system complexity while maintaining or improving positioning accuracy through sensor fusion.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
An apparatus for measuring, inspecting and/or processing objects is provided. The apparatus has a mobile platform for moving the apparatus through a region in space and a kinematic system attached to the mobile platform. An instrument head is fitted to the kinematic system. The apparatus furthermore comprises a controller configured to determine a first estimate of a pose of a target object on the basis of an image from a camera of the apparatus and a digital representation of the target object, to control the mobile platform on the basis of the first estimate, to move toward the target object, to determine a second estimate, in particular a more accurate estimate, of the pose of the target object on the basis of signals from at least one sensor of the apparatus and to control the kinematic system to position the instrument head on the target object on the basis of the second estimate. (Fig. 4)