Robotic Kitting Motion Planning for Variable Pick-and-Place Tasks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic systems struggle with fully automating kitting operations due to high variability in kits and objects, requiring extensive programming and human intervention, especially when objects are presented in random positions.
Innovation Solution
A computing system that captures images of objects and containers, estimates their poses using machine vision, and determines sequences for robotic motions, enabling multi-robot, multi-gripper systems to perform pick and place operations efficiently and flexibly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional robotic systems are used for kitting operations, then automation is partially achieved, but the system cannot adapt to high variability in kits and objects, requiring extensive programming and human intervention
Solution Approach 1:
The system performs self-programming by automatically generating motion plans based on real-time sensor data and digital twin models. The robot observes object positions and kit configurations, then autonomously computes pick-and-place sequences without human intervention, enabling adaptation to variable kits while maintaining simple operation interfaces
Solution Approach 2:
The system dynamically adjusts motion parameters, gripper forces, and trajectory points based on detected object properties and kit configurations. By changing operational parameters in real-time rather than reprogramming the entire system, the robot adapts to different kits and object arrangements efficiently
2Extent of automation
If robots are programmed to handle random object positions, then automation coverage increases, but the system requires extensive pre-programming and loses flexibility
Solution Approach 1:
The system performs preliminary observation and planning by capturing images of object positions and kit configurations before execution. The digital twin model pre-computes motion sequences based on observed states, enabling the robot to handle random positions without pre-programming while maintaining operational flexibility
Solution Approach 2:
The system continuously senses object positions and kit states during operation, comparing actual configurations against the digital twin model. This feedback loop enables real-time adjustments to motion plans, allowing the robot to adapt to unexpected variations while maintaining high automation coverage
3Productivity
If multiple robots are coordinated for complex kitting tasks, then productivity increases, but system complexity and coordination difficulty increase significantly
Solution Approach 1:
The digital twin model serves multiple functions simultaneously: it represents the physical kit configuration, computes motion plans for multiple robots, simulates collision-free trajectories, and validates final assembly. This universal modeling approach enables multi-robot coordination without requiring separate control systems for each robot, reducing overall system complexity
Data Source
AI summary
Fully flexible kitting processes can be automated by generating pick and place motions for multi-robot, multi-gripper, robotic systems.