Industries

How to achieve maximum safety with full regulatory compliance of ADAS and AV systems

Teaching a computer to drive as safe or safer than humans is more complex than anyone predicted. Advanced driver assistance systems (ADAS) and autonomous vehicle (AV) development require a continuous, multi-level, hybrid approach to verification and validation that spans the lifecycle of the vehicle. At the levels of the chip, the vehicle, and the environment, you must be able to verify that your design meets the requirements established at the beginning of the project and validate that your final product works as planned by using massive combinations of scenarios, parameters, and actors to achieve comprehensive coverage.

Watch the video to learn how to achieve maximum safety with full regulatory compliance of your ADAS and autonomous vehicle systems.

Validate ADAS and AV designs virtually at every level

Autonomous vehicles need to perceive and connect with the environment to ensure safety for everyone in and around the vehicle. Siemens Autonomous Vehicle Development solution enables a complex web of verification and validation (V&V) that spans all internal and external systems interacting with the vehicle. You can run a selection of tests in real-time across multiple levels of simulated environments, such as software-in-the-loop, hardware-in-the-loop, driver-in-the-loop, and vehicle-in-the-loop, each aimed at testing specific vehicle systems. Insights captured during validation can be fed into the design for continuous improvement.

Deliver trustworthy autonomy faster with closed-loop verification and validation

A systematic approach to continuously test, assess, and validate the functionality and safety of ADAS
features and AV as they evolve is essential to remain competitive. Verify that your final design meets the requirements and goals established at the beginning of the project and validate that your final product works as planned. To avoid the cost and expense of physical testing alone, you can leverage a hybrid of virtual and on-road testing to ensure safety and compliance in all situations.

Here’s how closed-loop virtual verification and validation across all systems levels continuously tracks and ensures compliance and performance from concept to delivery—

  1. Defining formal requirements and system architecture.
  2. Defining a digital twin of the world, containing static and dynamic objects and traffic agents, and of the vehicle.
  3. Configuration of the ego-vehicle and its occupants, including vehicle behavior.
  4. Sensor models ensure the closed-loop simulations of scenarios, where the virtual vehicles react on other traffic agents in a realistic way.
  5. Automatically generating test runs of millions of tests in a cloud or cluster setup.
  6. Running a selection of tests in real-time hardware-in-the-loop, driver-in-the-loop, and vehicle-in-the-loop setups.
  7. Executing test runs on the proving ground using driving robots and in-field tests on the road.
  8. This loop continuously runs up to the point where the car or system meets the requirements and can be certified for release on the market.

You’ll be able to evaluate systems across multiple levels of simulated environments, including software-in-the-loop, hardware-in-the-loop, driver-in-the-loop, and vehicle-in-the-loop before on-road testing begins. Identify any gaps that require further evaluation and feed optimizations back into designs to ensure complete coverage of requirements.

Achieve maximum active and passive safety with full regulatory compliance

Before autonomous vehicles can be deployed on the road, you must be confident they will perform flawlessly in every situation. With Siemens Autonomous Vehicle Development, you can leverage a closed-loop verification and validation process to—

  • Deliver safe autonomy and accelerate certification.
  • Verify requirements and validate that they are being met.
  • Reduce the risk of real-world failures by testing virtually against a growing database of critical scenarios.

Deliver trustworthy autonomy faster with closed-loop verification and validation from Siemens Autonomous Vehicle Development.

Kelsey Robuck
This article first appeared on the Siemens Digital Industries Software blog at https://blogs.stage.sw.siemens.com/automotive-transportation/2024/03/29/continually-verify-and-validate-adas-and-av-compliance-and-performance/