Hierarchical Avatar Control With Fewer Motion Sensors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current motion capture technologies require a large number of sensors to achieve realistic and faithful digital avatar movements, making them cumbersome and restrictive for users, and existing solutions like database-driven approaches are not generic or efficient.

Innovation Solution

A method for real-time control of a poly-articulated mass structure using a reduced number of motion sensors, where measurements are acquired, target points are determined, and optimization problems are solved to achieve hierarchical objectives, including balance and redundancy management, to generate commands for the structure.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a large number of sensors are used to capture movement, then the realism and fidelity of digital avatar movements is improved, but the device complexity and user restriction increase

Engineering Contradiction:
Improvemovement capture fidelityVSAvoidsensor installation complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The invention segments the control problem into hierarchical levels: high-level objectives (overall movement goals) and low-level objectives (specific joint configurations). This allows a reduced sensor set to capture essential movement data while the hierarchical controller reconstructs complete avatar movements by solving redundancy at each level, eliminating the need for extensive sensor coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The invention changes the control parameters from direct sensor-to-avatar mapping to hierarchical objective specifications. By formulating control as an optimization problem that minimizes a cost function based on hierarchical objectives, the system achieves realistic movements with fewer sensors, transforming the control approach from data-intensive to computation-intensive.

Inventive Principle:
Principle #35Parameter changes

2Ease of operation

If the number of sensors is reduced, then the ease of operation and user comfort is improved, but the measurement precision and movement fidelity deteriorate

Engineering Contradiction:
Improvesensor setup easeVSAvoidmovement data accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The hierarchical objective system acts as an intermediary between reduced sensor data and the full avatar model. Instead of directly mapping limited sensor readings to avatar joints, the system uses hierarchical objectives as intermediaries that guide the reconstruction process, ensuring movement fidelity is maintained despite reduced sensor input.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The invention replaces the mechanical/sensor-based solution (more sensors) with a computational solution (hierarchical optimization). By substituting physical measurement redundancy with computational reconstruction, the system maintains movement accuracy while reducing sensor requirements, improving ease of operation without sacrificing fidelity.

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

3Device complexity

If inverse kinematics techniques are used with reduced sensors, then the device complexity is reduced, but the reliability of movement representation deteriorates due to redundancy and balance problems

Engineering Contradiction:
Improvecontrol system simplicityVSAvoidmovement realism
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The invention introduces dynamics into the control system by formulating movement generation as a real-time optimization problem rather than a static inverse kinematics calculation. The cost function dynamically balances multiple hierarchical objectives, allowing the system to adapt to changing movement conditions and resolve redundancy and balance issues as they arise, improving movement realism.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The hierarchical objective framework provides continuous feedback control by evaluating how well current avatar configurations satisfy high-level and low-level objectives. The optimization process adjusts joint configurations in real-time based on this feedback, ensuring movement realism is maintained even with reduced sensors, unlike open-loop inverse kinematics approaches.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3484364B1Method and system for the hierarchical control in real time of an articulated structure utilising a limited number of movement sensors
Publication Date: 2022.03.09 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES
  • EP3484364B1 patent drawingFigure 1
  • EP3484364B1 patent drawingFigure 2
  • EP3484364B1 patent drawingFigure 3

AI summary

The invention relates to a method and a system for real-time control of an articulated mass structure from data representative of the movement of an object. The method comprises the following steps: - acquiring measurements made using a reduced number of movement sensors attached to the object; - determining target points from the measurements acquired; -determining objectives to fulfil for controlling the articulated mass structure, the objectives being arranged in a hierarchy and the highest level objective being the reaching of the target points by the articulated mass structure; and - determining, by real-time resolution of an optimisation problem, a command (CM) to be applied to the polyarticulated mass structure to fulfil the objectives arranged in a hierarchy.