Thireks develops stochastic traffic simulation technology that incorporates human misbehavior models to create millions of realistic driving scenarios for testing advanced driver assistance systems and automated driving solutions. This approach enables manufacturers to evaluate safety performance and compliance with SOTIF ISO 21448 standards, ultimately reducing accident rates and improving system reliability.
Funding
Funding not disclosed
Founders
Product
Problem
Testing advanced driver-assistance systems (ADAS) and autonomous driving (AD) solutions requires evaluating performance across a wide range of traffic scenarios, including rare and unexpected events. Traditional testing methods often rely on pre-defined scenarios that may not fully capture the complexities and randomness of real-world driving conditions, potentially leading to inadequate safety assessments. Ensuring compliance with Safety of the Intended Functionality (SOTIF) ISO 21448 standards requires a comprehensive approach to identifying and mitigating risks arising from design limitations and unforeseen situations.
Solution
Thireks provides stochastic traffic simulation technology that generates millions of realistic driving scenarios by incorporating human misbehavior models and other critical factors. This approach enables manufacturers to evaluate the safety performance of ADAS and AD systems under a multitude of conditions, including edge cases and potential accident scenarios. By simulating traffic with its inherent complexities and unexpected flaws, Thireks helps identify potential safety risks and improve system reliability. The simulations support compliance with SOTIF ISO 21448 standards by providing a comprehensive testing environment that accounts for unpredictable events, limitations in AI perception systems, and human interaction with automated systems.
Target Audience
The primary customers are automotive manufacturers, ADAS/AD system developers, and city planners focused on enhancing traffic safety and efficiency.
Features
- Stochastic traffic simulation generates millions of unique scenarios to ensure statistical significance in testing.
- Models incorporate human misbehavior, emergency reactions, and cognitive behavior, accounting for factors like age, gender, and skill.
- Simplifies complex traffic processes using stochastic distributions to model crucial aspects of driver behavior.
- Supports SOTIF ISO 21448 compliance by simulating unpredictable edge cases and system limitations.
- Aids in identifying and mitigating risks associated with AI perception systems and human-machine interaction.