Force-Guided Robot Motion Control for Adaptive Learning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current robotic systems lack the ability to enhance entertainment value through adaptive learning and interactive functionalities, particularly in humanoid robots, which limits their capability to acquire new functions and improve existing ones effectively.

Innovation Solution

A robotic system comprising a control unit, detection unit, and movement information deriving unit that allows the robot to detect external forces, derive movement direction and speed, and adjust its actions accordingly, enabling learning and improvement of functions such as running speed through user coaching and machine learning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If robots are equipped with basic movement functions, then they can perform simple tasks, but they lack the ability to learn and improve functions autonomously

Engineering Contradiction:
Improveability to learn and improve functionsVSAvoidcontrol system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements feedback mechanisms where the robot detects external forces applied to its body, derives movement information from these forces, and uses this information to adjust its movement actions. This closed-loop feedback enables the robot to learn from interactions and improve its functions autonomously without requiring complex reprogramming

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The robot performs self-learning by autonomously deriving movement directions and speeds from detected external forces, then adjusting its own control parameters based on this derived information. This self-service capability allows the robot to improve its movement functions through autonomous learning rather than external programming

Inventive Principle:
Principle #25Self-service

2Adaptability or versatility

If robots follow predetermined movement trajectories, then control is simple and stable, but they cannot adapt to external forces or learn new functions

Engineering Contradiction:
Improveadaptability to external forcesVSAvoidmovement stability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

The patent transitions from static predetermined trajectories to dynamic adaptive trajectories. The robot continuously detects external forces during movement, derives real-time movement information from these forces, and adjusts its trajectory dynamically. This dynamic approach maintains stability through controlled adaptation rather than rigid predetermined paths

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system uses feedback from force detection to continuously adjust movement trajectories. By detecting external forces and deriving movement information in real-time, the robot can adapt its trajectory while maintaining stability through controlled responses to external perturbations

Inventive Principle:
Principle #23Feedback

3Adaptability or versatility

If deep learning models are implemented in robots, then they can acquire new functions, but processing time and computational resources increase significantly

Engineering Contradiction:
Improvefunction acquisition capabilityVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent replaces complex computational deep learning models with a more efficient mechanical-sensing approach. Instead of using resource-intensive AI algorithms, the robot uses force detection and simple movement information derivation to achieve adaptive learning, significantly reducing processing time and computational resource requirements while maintaining function acquisition capability

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

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

The system enhances the entertainment value by allowing robots to learn and improve their functions autonomously, enabling them to perform tasks like running faster and participate in virtual competitions, thereby increasing user engagement and motivation.

Implementation Method 1

a detection unit configured to detect external force applied to the moving body

Methodology Applied
Scientific EffectForce detection: Force

Implementation Method 2

a movement information deriving unit configured to derive, based on the detected external force, a movement direction and movement speed of the moving body

Methodology Applied
Scientific EffectForce-to-motion conversion: Newton's Rings

Data Source

PatentUS11733705B2Moving body and moving body control method
Publication Date: 2023.08.22 SONY INTERACTIVE ENTERTAINMENT LLC
  • US11733705B2 patent drawing
  • US11733705B2 patent drawing
  • US11733705B2 patent drawing

AI summary

A control unit drives a drive source to move the moving body. A detection unit detects external force applied to the moving body. A movement information deriving unit derives, based on the detected external force, a movement direction and movement speed of the moving body. The control unit drives the drive source based on the movement direction and movement speed derived by the movement information deriving unit.