Unified Platform for Domain Adaptable Human Behavior Inference

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing human behavior inference frameworks are mostly generic and lack domain adaptability, failing to incorporate nuances of human sensing across multiple domains and are limited to specific sensor types like mobile phones, lacking interoperability with wearable and infrastructure sensors.

Innovation Solution

A unified platform for domain adaptable human behavior inference that integrates various sensor data formats, including wearable, infrastructure, and near-field sensors, using cross-sectional and longitudinal analysis techniques, along with domain knowledge and metadata, to provide low and high-level inferences.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a generic sensing framework is used, then ease of operation is improved, but adaptability to different human sensing domains deteriorates

Engineering Contradiction:
Improveease of operationVSAvoidadaptability
Core Design Contradiction:
Ease of operationVSAdaptability or versatility

Solution Approach 1:

The framework is designed to be universally applicable across multiple human sensing domains (e.g., elderly health care, ergonomics, worker safety) by incorporating domain-adaptable modules that can be configured for different applications. The system maintains a core generic architecture while enabling specialized functionality through configurable components that adapt to specific domain requirements.

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

2Device complexity

If independent separate entities limited to mobile phones are used, then device complexity is reduced, but interoperability with other sensors deteriorates

Engineering Contradiction:
Improvedevice complexityVSAvoidinteroperability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The framework merges multiple sensor types (mobile phone sensors, wearable sensors, infrastructure sensors, and near field sensors) into a unified sensing system. By combining these previously independent entities into an integrated architecture with standardized data processing pipelines, the system achieves interoperability while managing complexity through modular design patterns.

Inventive Principle:
Principle #5Merging (Combining)

3Adaptability or versatility

If domain-specific frameworks are developed for each application, then adaptability to specific domains is improved, but device complexity increases

Engineering Contradiction:
ImproveadaptabilityVSAvoiddevice complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The framework is segmented into modular components including domain-specific adaptation layers that can be independently configured. Each domain (health care, ergonomics, safety) has its own configurable module that interfaces with the core generic framework, allowing adaptability to specific domains without requiring complete redesign of the entire system and thus controlling overall complexity.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3561815B1A unified platform for domain adaptable human behaviour inference
Publication Date: 2025.07.16 TATA CONSULTANCY SERVICES LTD
  • EP3561815B1 patent drawingFigure 1
  • EP3561815B1 patent drawingFigure 2
  • EP3561815B1 patent drawingFigure 3

AI summary

This disclosure relates generally to a unified platform for domain adaptable human behaviour inference. The platform provides a unified, low level inference and high level inference of domain adaptable human behaviour inference. The low level inferences include cross-sectional analysis techniques to infer location, activity, physiology. Further the high inference that provide useful and actionable for longitudinal tracking, prediction and anomaly detection is performed based on several longitudinal analysis techniques that include welch analysis, cross-spectrum analysis, Feature of interest (FOI) identification and time-series clustering, autocorrelation-based distance estimation and exponential smoothing, seasonal and non-seasonal models identification, ARIMA modelling, Hidden Markov models, Long short term memory (LSTM) along with low level inference, human meta-data and application domain knowledge. Further the unified human behaviour inference can be obtained across multiple domains that include health, retail and transportation.