Robotic Placement Planning With Partial Pose Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional robotics planning, particularly placement planning, is tedious, time-consuming, and error-prone due to the need for extensive manual programming and over-specified constraints, which complicates the movement of robotic components and is susceptible to representation errors.
Innovation Solution
A system that simplifies placement planning by using only three inputs (x, y location, and object angle) on an insertion plane, constraining the search space, and automatically deriving constraints to generate a motion plan, allowing for faster convergence and reduced representation errors, especially for deformable objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional manual programming is used for placement planning, then the robotic system can perform tasks with detailed control, but the process becomes tedious, time-consuming, and error-prone
Solution Approach 1:
The system performs self-planning by automatically generating placement plans without requiring manual programming. The robotic system uses its own sensors and algorithms to determine placement parameters, eliminating the need for tedious manual programming while maintaining precise placement control
Solution Approach 2:
The patent replaces manual programming operations with an automated planning system that uses sensor data and algorithms to generate placement plans. This substitution eliminates the mechanical process of manual programming while achieving the same or better placement precision
2Manufacturing precision
If over-specified constraints are applied in placement planning, then the system can ensure precise placement, but the search space is unnecessarily limited and solution convergence becomes slower
Solution Approach 1:
The system applies only the necessary constraints for successful placement rather than over-specifying all possible constraints. By applying partial constraints (only those needed for the specific task), the system maintains placement precision while avoiding the negative effects of excessive constraint specification
Solution Approach 2:
The constraint application is dynamic and adaptive to the specific placement task. The system adjusts the level and type of constraints based on the object being placed and the target location, rather than applying fixed over-specified constraints to all cases. This dynamic approach optimizes both precision and convergence speed
3Adaptability or versatility
If full 6DOF planning is performed, then the system can handle complex placement scenarios, but the computational complexity increases and the system becomes more susceptible to representation errors
Solution Approach 1:
The patent segments the placement planning task into distinct components: determining placement parameters from sensor data, generating the placement plan, and executing the placement. This segmentation allows the system to handle complex scenarios reliably by breaking down the 6DOF planning into manageable steps, reducing the accumulation of representation errors
Solution Approach 2:
The system introduces an intermediate planning stage that acts as a mediator between sensor input and execution. This planning intermediary processes the 6DOF information in a structured way, reducing susceptibility to representation errors by validating and refining the data before final execution
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on computer storage media, for performing planning for robotic placement tasks. One of the methods includes receiving a request for a robot to perform a placement task with an object, wherein the placement task has an associated insertion plane, wherein the request specifies a location on the insertion plane and an orientation angle for the object. A plan is generated for the robot including generating a rotational motion to transition the gripped object from an initial orientation to an orientation in which a principal vector is aligned with the insertion plane and the object is aligned with the orientation angle of the request.