Digital Twin Cognitive Models for Human Behavior Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current behavior estimation tools for companion robots and similar applications face challenges in reliably mimicking human behavior, particularly in tasks that require reasoning about unobservable factors, which existing designs struggle to replicate.
Innovation Solution
A computer-implemented method and apparatus that generate a digital twin of a person by analyzing digital memories and their associations, using personalized cognitive models to emulate cognitive skills, which are then incorporated into a behavior tool to influence its estimation of the person's behavior in given situations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional behavior estimation methods are used, then the behavior tool can operate with simpler architecture, but it cannot reliably estimate human behavior especially for unobservable factors
Solution Approach 1:
The patent creates a digital twin that copies the cognitive architecture of a human person, including personalized cognitive models for reasoning, decision-making, and memory. This virtual replica processes information the same way a human would, enabling reliable estimation of unobservable behavioral factors without requiring complex physical sensors or actuators.
Solution Approach 2:
The digital twin serves as an intermediary between the behavior tool and the human person. Instead of directly observing and interpreting complex human behavior, the system creates a simplified virtual model that processes observable data through human-like cognitive processes to produce accurate behavior estimates.
2Measurement precision
If personalized cognitive models are generated from digital memories, then the behavior estimation becomes more accurate, but the data processing and model generation process becomes more complex
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing observational data in digital memories before they are needed for behavior estimation. Cognitive models are trained in advance on these pre-organized memories, so that when behavior estimation is required, the models can quickly process information without re-processing raw data, reducing real-time computational complexity.
Solution Approach 2:
The cognitive system is segmented into separate personalized cognitive models, each responsible for specific functions such as reasoning, decision-making, or memory retrieval. This modular architecture allows the complex data processing task to be divided into manageable segments that can be processed independently and in parallel.
3Adaptability or versatility
If digital memories and associations are analyzed to create digital twin, then the behavior tool can replicate human reasoning, but the computational resources and processing time increase
Solution Approach 1:
The digital twin's cognitive models dynamically adjust their processing based on the complexity of the situation. For simple situations, the models use cached digital memories and associations to quickly generate responses. For complex situations requiring human-like reasoning, the models activate more comprehensive processing networks, optimizing energy consumption by matching computational effort to task requirements.
Data Source
AI summary
A computer implemented technique for developing a behaviour tool to estimate a behaviour of a particular person in response to a given situation, comprises maintaining, in a storage device, digital memories and a record of associations between the digital memories, where a given digital memory is generated in response to a given event associated with the particular person and is determined from analysis of multiple items of data associated with the given event, including at least items of personal data derived from signals gathered from a plurality of sensors used to monitor the particular person. Processing circuitry is then employed to analyse the digital memories and the record of associations between the digital memories, in order to generate a digital twin of the particular person comprising one or more personalised cognitive models, each personalised cognitive model being arranged to emulate an associated cognitive skill of the particular person. At least one of the one or more personalised cognitive models forming the digital twin is then output for incorporation within the behaviour tool so as to cause the estimated behaviour of the particular person in response to the given situation to be influenced, at least in part, by model output data generated by the at least one of the one or more personalised cognitive models.