Point Cloud Alignment Detection for Stacking Equipment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Material handling equipment faces challenges in accurately aligning stacked objects due to external environment factors and equipment errors, affecting operation safety during the stacking process.
Innovation Solution
A method involving a controller that uses sensors to acquire point clouds of stacking objects, extracts specific regions, matches them with template point clouds to determine poses, and compares differences to assess alignment, ensuring precise stacking alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If material handling equipment stacks objects automatically without alignment detection, then productivity is improved, but manufacturing precision deteriorates due to accumulation of positioning errors
Solution Approach 1:
The system uses sensors to detect the actual positions of stacking objects in real-time, feeds this information back to the controller, and dynamically adjusts positioning to compensate for deviations. This closed-loop feedback mechanism maintains high stacking speed while ensuring precise alignment by continuously correcting positioning errors as they occur.
Solution Approach 2:
The patent replaces traditional mechanical alignment methods with vision-based detection and sensor-based positioning systems. By using cameras, lasers, or other sensing devices to detect object positions and calculate alignment states, the system achieves higher precision without mechanical contact, enabling fast automated stacking while maintaining accurate alignment through software-based correction.
2Manufacturing precision
If alignment detection is performed using traditional methods, then manufacturing precision is improved, but device complexity increases due to additional sensors and processing requirements
Solution Approach 1:
The system integrates multiple functions into a single controller that performs both stacking control and alignment detection. The controller uses sensor data not only for positioning the stacking mechanism but also for detecting alignment states between objects. This multi-functional approach achieves precise alignment without requiring separate dedicated alignment devices, thereby reducing overall system complexity.
Solution Approach 2:
The stacking system uses its own sensor data and control mechanisms to perform self-alignment. The controller analyzes the relative positions of objects using the same sensing system that guides the stacking operation, enabling the system to automatically detect and correct alignment issues without external intervention or additional specialized equipment, thus maintaining simplicity while achieving high precision.
3Measurement precision
If point cloud matching is performed on entire objects, then measurement precision is improved, but loss of time increases due to large data processing requirements
Solution Approach 1:
The system extracts only the necessary feature points or key regions from the complete point cloud data of stacking objects, such as corners, edges, or specific reference markers. By processing only these extracted features rather than the entire point cloud, the system maintains accurate pose detection and alignment measurement while dramatically reducing data processing time and computational load.
Solution Approach 2:
The patent segments the point cloud data into multiple regions or features of interest, processing each segment independently to determine object pose and alignment. This segmentation approach allows parallel processing of different object features, maintaining measurement precision by focusing on critical alignment points while reducing overall processing time through divided computational tasks that can be executed simultaneously or in optimized sequences.
Data Source
AI summary
A method for determining an alignment state includes: acquiring, by using a sensor, target point clouds of a first and second stacking object; extracting, from the target point clouds, a first point cloud of a first target region of the first stacking object and a second point cloud of a second target region of the second stacking object; matching the first point cloud with a first template point cloud, to obtain a pose of the first stacking object, and matching the second point cloud with a second template point cloud, to obtain a pose of the second stacking object; and determining a difference between the poses of the first and second stacking object, and comparing the difference with a threshold, to determine an alignment state between the first and second stacking object. The present disclosure is used to implement alignment between a first and second stacking object during stacking.