Neural Network Programming via Distributed Observer Signal Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computing models in automotive safety applications are inadequate to process the increasing volumes of diverse and unstructured data, such as video and images, and require a new approach for programming cognitive computing systems to enhance their learning and decision-making capabilities.
Innovation Solution
Integrating observer neural network computers into active safety systems of multiple vehicles to observe signals from forward-facing cameras and driver action monitoring devices, correlating and combining these signals to program a target neural network computer, allowing it to learn from the experiences of numerous drivers and enhance its self-learning capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional computing models are used to process automotive safety data, then processing speed is maintained, but the ability to handle diverse unstructured data (video, images, symbols) is insufficient
Solution Approach 1:
The patent replaces traditional mechanical computing models with cognitive computing systems based on neural networks. These neural network computers can process unstructured data (video, images, symbols) by mimicking human cognitive processing, thereby improving adaptability to diverse data types while maintaining acceptable processing speeds through parallel processing capabilities inherent in neural network architectures.
2Adaptability or versatility
If cognitive computing systems are programmed with pre-programmed instructions, then processing reliability is maintained, but the ability to learn from experience and adapt is limited
Solution Approach 1:
The patent implements self-service through observer neural network computers that automatically learn from real-world driving experiences without requiring manual reprogramming. These observers collect data from multiple vehicles, identify patterns, and autonomously update the target neural network computer's programming, thereby improving learning capability while reducing programming complexity through automated self-learning mechanisms.
Solution Approach 2:
The patent incorporates feedback loops where observer neural network computers continuously monitor real-world outcomes of safety system decisions, compare expected versus actual results, and use this feedback to refine and update the target neural network computer's programming. This feedback mechanism enables continuous improvement of learning capabilities while maintaining system reliability.
3Measurement precision
If observer neural network computers are integrated into multiple vehicles to collect data, then learning accuracy is improved, but data processing time and system complexity increase
Solution Approach 1:
The patent applies preliminary action by having observer neural network computers continuously collect and pre-process data in the background during normal vehicle operation. This data collection and initial processing occurs asynchronously without interrupting vehicle operations, thereby improving learning accuracy through extensive data gathering while minimizing impact on data processing time through parallel execution.
Solution Approach 2:
The patent merges data from multiple observer neural network computers across different vehicles into a unified training dataset. By combining correlated signals from numerous observers, the system improves learning accuracy through diverse real-world examples while efficiently processing this aggregated data using distributed computing resources and optimized neural network training algorithms.
Data Source
AI summary
A method is disclosed for programming a target neural network computer for use in cognitive computing systems in automotive safety applications. An observer neural network computer is integrated into active safety systems of a plurality of vehicles to observe signals. Each respective observer neural network computer is arranged to observe signals from a forward facing camera and signals from a driver action monitor of its respective vehicle, to process the observed signals from the forward facing camera of its respective vehicle and correlate them with the observed signals from the driver action monitor of its respective vehicle. The correlated signals from the plurality of observer neural network computers are combined, and the target neural network computer is programmed for use in cognitive computing systems in automotive safety applications based on said combined correlated signals.