Autonomous Drive Robustness via Sense-Act Map
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems face challenges in robustly and safely handling the multitude of vehicle systems, operational domains, and loads, similar to human drivers, due to the complexity of design parameters and noise interactions, which existing big data approaches like AI and Machine Learning are unable to effectively manage, leading to inefficiencies and high computational demands.
Innovation Solution
A holistic deterministic approach using analytical statistical methods, including transfer functions, weighted optimization, and data mining, to create a general robust Sense-Act-interdependency relations map that enables efficient decision-making with global situational awareness, closed-loop control, and fail-safe management, without the need for complex code.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If big data approaches (AI and Machine Learning) are utilized to manage autonomous drive design, then data processing capability is improved, but computational resource demands and complexity increase significantly
Solution Approach 1:
The patent replaces complex AI/ML computational systems with a deterministic analytical statistical approach using transfer functions. This substitution reduces computational resource demands while maintaining data processing capability by using mathematical models that describe system behavior without requiring extensive training data or computational power.
Solution Approach 2:
The patent changes the fundamental parameters of the approach from data-intensive AI/ML methods to parameter-based analytical statistical methods. By using transfer functions with defined parameters that capture system behavior, the solution achieves comparable productivity with significantly reduced computational complexity.
2Reliability
If a holistic deterministic approach using analytical statistical methods is used, then robustness and safety are improved, but design and implementation complexity increases
Solution Approach 1:
The patent segments the complex autonomous driving system into modular transfer functions, each representing a specific subsystem or functional relationship. This segmentation allows the holistic deterministic approach to be broken down into manageable, independently analyzable components, reducing overall design and implementation complexity while maintaining robustness and safety.
3Adaptability or versatility
If the number of load cases is increased to cover all operational domains, then coverage and completeness are improved, but the problem becomes exponentially more difficult to solve
Solution Approach 1:
The patent creates universal transfer functions that can handle multiple operational domains and load cases simultaneously. Instead of developing separate models for each scenario, the transfer functions are designed to be universally applicable across diverse conditions, maintaining coverage and completeness while avoiding exponential problem difficulty.
Data Source
AI summary
A method, a non-transitory computer-readable medium, and a system of providing an autonomous driving (AD) generally valid robust Sense-Act-function input-output Sense-Act-interdependency relations knowledge-based map that is used as look-up table instructions to implement general Sense-Act-control capabilities of an AD system in an AD vehicle for general robust solution decision making with global situational awareness, closed-loop control, noise and system degradation-tolerance, and fail-safe management and without large and complex code that is difficult, bug-sensitive, and both time and resource consuming to both develop and execute.