Robotic Object Placement Using Expanded Interference Regions

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for checking interference in the motion path of robot models require high implementation costs and longer processing times due to complex model generation, making it inefficient for robotic hands to grasp and handle objects without colliding with peripheral obstacles.

Innovation Solution

An object handling device that calculates an expanded interference region to determine optimal placement and release positions for objects within a container, using a combination of interference regions from the hand and the object, while avoiding collisions with surrounding obstacles, allowing for high-speed and feasible handling.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If methods for checking interference in the motion path of a robot model are used, then interference detection capability is improved, but implementation cost increases and processing time lengthens

Engineering Contradiction:
Improveinterference detection capabilityVSAvoidimplementation cost
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent creates a simplified two-dimensional interference region map as a copy of the three-dimensional environment, viewed from above. This 2D representation captures essential interference information without requiring complex 3D models, thereby reducing implementation cost while maintaining interference detection capability. The control device generates this 2D map by processing sensor data to identify regions where the robotic hand may interfere with objects or obstacles.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent extracts only the necessary interference information from the complex environment and represents it in a simplified 2D format. By taking out only the relevant interference regions and obstacles from the full 3D scene and representing them in a top-down 2D view, the system achieves efficient interference detection without the computational burden of complete 3D modeling.

Inventive Principle:
Principle #2Taking out (Extraction)

2Reliability

If methods for checking interference in the motion path of a robot model are used, then interference detection capability is improved, but processing time increases

Engineering Contradiction:
Improveinterference detection capabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The 2D interference region map serves as a computationally efficient copy of the environment that can be processed quickly. By representing interference zones and obstacles in a simplified 2D top-down view rather than full 3D models, the control device can generate and query interference information in real-time, significantly reducing processing time while maintaining detection accuracy.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent segments the environment into discrete interference regions and non-interference regions in the 2D map. This segmentation allows the control device to efficiently query and process only relevant regions during motion planning, avoiding the need to analyze entire 3D models and thereby reducing processing time.

Inventive Principle:
Principle #1Segmentation

3Reliability

If the robotic hand and object are positioned to avoid obstacles, then collision-free handling is achieved, but the complexity of position calculation increases

Engineering Contradiction:
Improvecollision-free handlingVSAvoidposition calculation complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent transforms the three-dimensional position calculation problem into a two-dimensional problem by using a top-down view of the interference region map. The control device determines whether placement positions are in non-interference regions by comparing 2D coordinates on the map, which simplifies the calculation complexity while ensuring collision-free handling. This dimensional reduction allows for efficient determination of safe placement positions without complex 3D geometric computations.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS11498210B2Object handling device and computer program product
Publication Date: 2022.11.15 KK TOSHIBA
  • US11498210B2 patent drawing
  • US11498210B2 patent drawing
  • US11498210B2 patent drawing

AI summary

According to one embodiment, an object handling device causes a hand to grasp, convey, and place an object in a container. The object handling device includes one or more processors configured to calculate an expanded interference region by combining an interference region of the hand interfering with another object and an interference region of the grasped object interfering with the another object; calculate at least a placement position between the placement position and a placement posture of the object in the container in accordance with information on an obstacle in the container and the expanded interference region; and calculate a release position at which the hand releases the object, on the basis of the placement position of the object.