Flexible Robot Gripper Pose Estimation Using Tactile Curvature Sensing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Flexible robotic end effectors face challenges in accurately determining the pose of objects due to their compliance, which makes precise placement difficult in pick-and-drop scenarios, especially when dealing with oddly shaped or unknown objects.

Innovation Solution

The use of tactile and curvature sensors integrated into flexible end effectors to detect points of contact and deformations, allowing for the calculation of relative transformations and determination of object pose through a pose determination algorithm.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If flexible end effectors are used to grasp oddly shaped objects, then the ability to grasp difficult objects is improved, but the precision of pose determination deteriorates

Engineering Contradiction:
Improveability to grasp oddly shaped objectsVSAvoidpose determination precision
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The flexible end effector is divided into multiple segments along its length, with each segment equipped with tactile and curvature sensors. This segmentation allows the system to capture localized deformation and contact information at discrete points, which are then integrated to determine the overall pose of the grasped object, resolving the precision issue while maintaining flexibility.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent replaces traditional mechanical pose determination methods with sensor-based measurement systems. Tactile sensors detect contact forces and curvature sensors measure bending at each segment, substituting mechanical assumptions about rigid body behavior with direct sensory measurement of the flexible structure's actual state during grasping.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If tactile and curvature sensors are integrated into flexible end effectors, then pose determination capability is improved, but device complexity increases

Engineering Contradiction:
Improvepose determination capabilityVSAvoidsensor integration complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The flexible end effector structure serves multiple functions: it provides the grasping capability through its flexibility while simultaneously serving as the mounting structure for sensors. The segments that enable compliance also host the tactile and curvature sensors, eliminating the need for separate sensor mounting mechanisms and reducing overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent merges the structural components of the flexible end effector with the sensor integration architecture. The flexible segments themselves become the sensor platforms, combining mechanical function and sensing function into a unified structure, thereby reducing the number of separate components and simplifying the overall device.

Inventive Principle:
Principle #5Merging (Combining)

3Measurement precision

If continuous data stream from sensors is processed, then pose estimation accuracy is improved, but computational requirements increase

Engineering Contradiction:
Improvepose estimation accuracyVSAvoidcomputational energy requirements
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary processing of sensor data by segmenting the flexible end effector into discrete sections and assigning coordinate frames to each segment before full pose calculation. This pre-organization of data structures and coordinate transformations reduces the computational complexity of the subsequent pose estimation algorithm, lowering energy requirements while maintaining accuracy.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables precise estimation of object pose, ensuring accurate placement by converting the data from sensors into continuous streams for precise robotic movements.

Implementation Method 1

one or more tactile sensors positioned adjacent to each one of the one or more flexible fingers, the one or more tactile sensors configured to sense a location of one or more deformations of the flexible internal side member caused by a contact between the flexible end effector and the object held by the flexible end effector

Methodology Applied
Scientific EffectDeformation: Deformation

Implementation Method 2

one or more curvature sensors positioned to sense a curvature of each one of the one or more flexible fingers and generate curvature data corresponding to the curvature

Methodology Applied
Scientific EffectCurvature sensing:

Data Source

PatentEP4037878B1Systems and methods for determining pose of objects held by flexible end effectors
Publication Date: 2023.12.20 TOYOTA RESEARCH INSTITUTE INC
  • EP4037878B1 patent drawingFigure 1
  • EP4037878B1 patent drawingFigure 2A~2B
  • EP4037878B1 patent drawingFigure 3

AI summary

Systems and methods for determining a pose of an object held by a flexible end effector of a robot are disclosed. A method of determining a pose of the object includes receiving tactile data from tactile sensors, receiving curvature data from curvature sensors, determining a plurality of segments of the flexible end effector from the curvature data, assigning a frame to each segment, determining a location of each point of contact between the object and the flexible end effector from the tactile data, calculating a set of relative transformations and determining a location of each point relative to one of the frames, generating continuous data from the determined location of each point, and providing the continuous data to a pose determination algorithm that uses the continuous data to determine the pose of the object.