Autonomous Bin Picking with Sensor Fusion and Grasp Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current bin picking technologies face challenges in combining object detection and grasp detection outputs effectively for autonomous robots, lacking a systematic approach to decision-making, especially when dealing with mixed object types and complex environments, leading to inefficiencies and inaccuracies.
Innovation Solution
The implementation of high-level sensor fusion and multi-criteria decision-making methodologies to combine outputs from object detection and grasp detection modules, computing attributes for grasping alternatives, and ranking them to select the optimal grasping action for execution by a robot.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple object detection and grasp detection modules are used to improve accuracy, then the reliability of bin picking increases, but the device complexity increases
Solution Approach 1:
The system divides the bin picking task into separate functional modules: object detection modules that identify objects in the bin, and grasp detection modules that determine grasping actions. Each module operates independently and produces separate outputs that are later integrated, allowing the system to achieve high reliability through modular design while managing complexity through clear separation of concerns.
Solution Approach 2:
A decision-making module acts as an intermediary that integrates the outputs from multiple object detection and grasp detection modules. This mediator synthesizes information from various sources, computes attributes for different grasping alternatives, and selects the optimal action, thereby coordinating the complex interactions between multiple modules without requiring direct integration of all components.
2Measurement precision
If high-level sensor fusion and multi-criteria decision making are implemented to improve bin picking accuracy, then measurement precision increases, but device complexity increases
Solution Approach 1:
The decision-making module evaluates multiple criteria parameters for each grasping alternative, including success probability, ease of execution, and task relevance. By changing from single-criterion to multi-criteria evaluation, the system achieves higher measurement precision in selecting grasping actions. The complexity is managed by systematically organizing these parameters into a structured decision-making framework that processes multiple attributes simultaneously.
Data Source
AI summary
In described embodiments of method for executing autonomous bin picking, a physical environment comprising a bin containing a plurality of objects is perceived by one or more sensors. Multiple artificial intelligence (AI) modules feed from the sensors to compute grasping alternatives, and in some embodiments, detected objects of interest. Grasping alternatives and their attributes are computed based on the outputs of the AI modules in a high-level sensor fusion (HLSF) module. A multi-criteria decision making (MCDM) module is used to rank the grasping alternatives and select the one that maximizes the application utility while satisfying specified constraints.