Vehicle Following Distance Detection with Virtual Tailgating Scenarios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining vehicle following distance are limited in the context of determining vehicle following distance are inadequate for creating training data for training an artificial intelligence to predict a distance between two vehicles.
Innovation Solution
A method for creating training data for an artificial intelligence to predict a distance between two vehicles, comprising accessing respective parameter data, simulating and rendering images, and storing them in a virtual environment, and a system for determining the same.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If real-world vehicle data is used for training AI, then the training data reflects actual driving conditions, but it is difficult to obtain sufficient diverse scenarios including rare dangerous situations
Solution Approach 1:
The patent creates virtual copies of real-world driving scenarios through simulated environments. Instead of relying solely on real-world data collection, the system generates synthetic training data by copying and reproducing driving conditions, vehicle positions, and environmental factors in a virtual setting. This allows comprehensive coverage of all possible scenarios including rare dangerous situations without the limitations of real-world data collection.
Solution Approach 2:
The patent performs preliminary generation of training data before actual AI training occurs. By pre-generating diverse driving scenarios, vehicle positions, and environmental conditions in the virtual environment, the system prepares comprehensive training datasets in advance. This preliminary action ensures that when AI training begins, all necessary scenario variations are already available, including edge cases and dangerous situations that would be difficult to capture in real-world collection.
2Reliability
If more diverse training scenarios are generated, then the AI model becomes more robust, but the data collection and processing time increases
Solution Approach 1:
The patent replaces mechanical real-world data collection methods with computational virtual environment simulation. Instead of physically moving vehicles and cameras to capture diverse scenarios, the system uses software-based virtual environments to generate identical diversity through code. This substitution dramatically reduces the time required to generate training data while maintaining comprehensive scenario coverage, as virtual scenario generation is much faster than physical data collection.
3Productivity
If virtual environment simulation is used to generate training data, then diverse scenarios can be created efficiently, but the complexity of the simulation system increases
Solution Approach 1:
The patent introduces a virtual environment as an intermediary layer between real-world driving conditions and AI training data. This intermediary simulation system translates complex real-world physics and scenarios into simplified virtual representations that are easier to generate and manipulate. The virtual environment acts as a mediator that handles the complexity of scenario generation internally, presenting simplified training data to the AI system while maintaining efficiency in data production.
Data Source
AI summary
Systems, methods, models, and training data for models are discussed, for determining vehicle positioning, and in particular identifying tailgating. Simulated training images showing vehicles following other vehicles, under various conditions, are generated using a virtual environment. Models are trained to determine following distance between two vehicles. Trained models are used to in detection of tailgating, based on determined distance between two vehicles. Results of tailgating are output to warn a driver, or to provide a report on driver behavior.