Rehabilitation Robot Joint Control Without Torque Sensors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing rehabilitation training robots lack the ability to dynamically adjust driving force based on the interacting force between the rehabilitation trainee and the robot, limiting their effectiveness in considering subtle changes during rehabilitation training.

Innovation Solution

A method and device that analyze the interacting force between the rehabilitation trainee and the robot, classify its characteristics, and control the driving force of the joint part to ensure movement along a preset trajectory, using position measurement to adjust driving force in real time without a torque sensor.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If the rehabilitation training robot uses a simple structure to perform specific operations repeatedly, then the device complexity is reduced, but the adaptability to different rehabilitation conditions and subtle changes during training deteriorates

Engineering Contradiction:
Improvestructure complexityVSAvoidadaptability to rehabilitation conditions
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamic adjustment of driving force based on real-time classification of interacting force characteristics. The driving part dynamically modifies its output according to the trainee's actual force interactions, transitioning from static pre-programmed operations to adaptive real-time control. This resolves the contradiction by enabling complex adaptive behavior through dynamic control rather than complex mechanical structure.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system changes operational parameters (driving force magnitude and direction) based on classified interacting force characteristics. By adjusting driving force parameters in real-time according to measured interactions, the system achieves adaptability without requiring complex mechanical reconfiguration, thus maintaining simple structure while improving versatility.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If the rehabilitation training robot actively assists rehabilitation by considering the rehabilitation status of the trainee, then the adaptability is improved, but the device complexity and manufacturing cost increase

Engineering Contradiction:
Improveadaptability to rehabilitation statusVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent replaces complex mechanical sensing systems with a computational approach. Instead of using torque sensors or force sensors to directly measure interactions, the system calculates interacting force characteristics from position and motion data. This substitution of mechanical sensing with computational analysis reduces device complexity while maintaining adaptability to rehabilitation status.

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

Solution Approach 2:

The patent introduces a classification module as an intermediary between force measurement and control execution. This intermediary classifies interacting force characteristics into discrete categories, simplifying the control logic and reducing the complexity of the overall system while still enabling adaptive assistance based on rehabilitation status.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If the rehabilitation training robot does not sufficiently consider subtle changes during training such as interacting force, then the device complexity is reduced, but the measurement precision and rehabilitation effectiveness deteriorate

Engineering Contradiction:
Improvecontrol system complexityVSAvoiddetection of interacting force
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent implements feedback control by continuously measuring interacting force characteristics and using this information to adjust driving force in real-time. The classification of interacting force characteristics provides feedback signals that enable the system to respond to subtle changes during training, improving measurement precision without proportionally increasing control system complexity through discrete feedback categories.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4691444A1Rehabilitation training method and rehabilitation training device
Publication Date: 2026.02.11 P&S ROBOTICS CO LTD
  • EP4691444A1 patent drawingFigure 1
  • EP4691444A1 patent drawingFigure 2(a)~2(b)
  • EP4691444A1 patent drawingFigure 3

AI summary

The present invention relates to a rehabilitation training method and a rehabilitation training device, which control provided driving force of a driving part provided at a joint part, so as to adjust the displacement and speed of the joint part. The rehabilitation training method may comprise steps in which: a rehabilitation trainee wears a rehabilitation training robot having at least one joint part, and interacting force between the rehabilitation trainee and the rehabilitation training robot is obtained while the joint part is driven by the driving part; the characteristics of the interacting force are classified by analyzing the interacting force; and the provided driving force of the driving part provided at the joint part is controlled according to the characteristics of the interacting force.