Vision System for Teat Detection Using Dynamic Algorithm Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dairy milking operations face inefficiencies due to the inability of existing systems to accurately and efficiently position and maneuver robotic arms around dairy livestock, which move and have varying teat positions, and the need to avoid obstacles like tails.
Innovation Solution
A vision system equipped with a robotic arm, memory devices, and a processor that uses 3D images and profile signals to detect and compensate for leg, teat, and tail movements, allowing real-time adjustments and successful teat cup attachment, regardless of cow position or tail presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hard coded movements and positions are used for robotic arm operations, then the system is simple to implement, but it cannot adapt to moving livestock and varying teat positions
Solution Approach 1:
The system transitions from static hard-coded positions to dynamic real-time positioning by using vision systems to continuously track livestock movement and teat positions, allowing the robotic arm to adapt its trajectory and attachment points dynamically during operation
Solution Approach 2:
The patent replaces mechanical positioning systems with vision-based detection and control systems. Cameras and image processing algorithms substitute for physical guides and hard-coded mechanical paths, enabling the robotic arm to locate and attach to teats without pre-programmed positions
2Reliability
If the robotic arm operates without real-time detection capabilities, then the system is simpler, but it cannot avoid obstacles like tails or compensate for movement
Solution Approach 1:
The vision system provides continuous feedback on livestock position, tail location, and teat coordinates to the robotic arm controller. This real-time feedback loop enables the system to adjust its operations dynamically, avoiding obstacles and compensating for movement to maintain high attachment success rates
Solution Approach 2:
The vision system acts as an intermediary between the robotic arm and the livestock. It processes visual information about the environment and translates it into actionable guidance for the robotic arm, mediating the interaction between the automated system and the moving animal
3Reliability
If multiple attach algorithms are tested to find the best method, then attachment success improves, but the time required for each milking operation increases
Solution Approach 1:
The system performs preliminary testing of multiple attach algorithms during setup or maintenance periods to build a knowledge base of effective methods. During actual milking operations, it retrieves pre-determined successful algorithms for that specific livestock, avoiding time-consuming real-time testing while maintaining high attachment success
Solution Approach 2:
The robotic system automatically selects and executes the most appropriate attach algorithm based on real-time vision data and historical performance. It self-optimizes by learning from past operations and automatically adjusting its approach without requiring manual intervention or time-consuming trial-and-error during milking
Data Source
AI summary
A system includes a robotic arm, memory, and a processor. The processor is configured to access first and second success counters stored in the memory. The first success counter is associated with a first attach algorithm and the second success counter is associated with a second attach algorithm. The processor is further configured to determine whether the first success counter is greater than the second success counter. The processor is further configured to execute the first attach algorithm if the first success counter is greater than the second success counter, and to execute the second attach algorithm if the first success counter is not greater than the second success counter. The processor is further configured to increment the first success counter if the first attach algorithm is successful, and to increment the second success counter if the second attach algorithm is successful.