Analyzing Critical ADAS/AD Scenarios With AVL Scenario Simulator™

  • Blog

To determine if an automated driving function is safe, billions of test kilometers would be required. Physical testing and real-world prototypes simply cannot efficiently handle such a massive test volume. Virtualization offers a more sustainable option that can manage the enormous test volume required at a much lower cost.

However, when running millions of tests, it is important to aggregate the results, find critical failed tests and analyze them in detail. One of the main questions for an ADAS/AD function developer is: "How well does my ADAS/AD function behave in critical scenarios?" Software developers typically use CI/CD systems to answer very similar questions. Whenever they change the code, they want to know how many of the automated tests the code change breaks.

AVL Scenario Simulator™ is a continuous integration and testing platform for ADAS/AD software and provides simulation models and tool interfaces to create a virtual world for reproducible scenario-based tests based on ASAM OpenSCENARIO® standard.

Simulation Blog - Analyzing Critical ADAS/AD Scenarios With AVL Scenario Simulator™

In AVL Scenario Simulator™, analysis takes place on two levels: The run history provides an aggregated view on all executed simulations based on test parameters and KPIs/safety metrics. This allows for a quick high-level comparison of different ADAS/AD software versions. It also allows to cluster and filter failed tests for an efficient detail analysis of individual tests.

gl_ast-blog-analyzing-critical-conditions-01-07-2024-fhd.png
Figure 1: A typical Scenario Simulator model. It consists of the scenario engine (esmini), a simple vehicle model, a simple radar sensor model and the ADAS function in the middle.

The Scenario Replay feature enables a 3D playback of recorded ASAM OSI® data as well as timeseries data. This allows analysis from all angles and different sensor data streams to investigate the exact cause of a failed test and provides important debugging information for the ADAS/AD developers. Based on this information, they can improve the handling of the safety-critical situation in the next version of the ADAS/AD software.

In the following example, the unit under test is the ADAS/AD function. The function developer changes the ADAS/AD function and wants to see how successful the function behaves. The success of such a function is measured by common KPIs like time to collision during the simulation of a scenario. Time to collision is defined as the time delay remaining between the time instant considered and the time of the collision between the vehicle and any other vehicle, irrespective any alteration of speeds and directions during that time delay. Obviously, this should never be below a certain threshold.

gl-ast-blog-analyzing-critical-conditions-02-07-2024
Figure 2: Logical and concrete scenarios are represented as case sets and cases.

For the test, we used what is called a logical scenario. A logical scenario defines the parameters of the scenario. By applying concrete parameter values, concrete scenarios can be created. In the screenshot below you can see that we use the usual SDT case sets to represent logical scenarios and cases to represent specific scenarios.

gl-ast-blog-analyzing-critical-conditions-03-07-2024
Figure 3: The run overview of AVL Scenario Simulator™

The simulation dashboard gives an aggregated overview of all performed simulations for the project. It enables software developers and validation engineers to analyze simulation results, compare to older versions and deep dive into single failed tests with the 3D scenario replay tool.

In the dashboard view, the user can start runs and see the progress of the simulations. A green tile in this window indicates that the KPIs of a particular scenario are within the allowed ranges, while a red tile indicates that some KPIs violate the acceptance criteria. Blue tiles represent running simulations and gray tiles represent pending jobs.

By clicking on individual tiles, users can access more detailed information about each case on the right side of the screen (such as parameter and KPI values).

gl-ast-blog-analyzing-critical-conditions-04-07-2024
Figure 4: The detailed case view of AVL Scenario Simulator™

For a more detailed analysis of what actually happened in a scenario, the user can click the Replay button and review the recorded simulation in the Scenario Replay.

gl-ast-blog-analyzing-critical-conditions-05-07-2024
Figure 5: The Scenario Replay showing three recorded OSI streams of the simulation.

These dashboard views and the Scenario Replay are created using web technology to be already prepared for the upcoming task of running all these simulations in the cloud and just providing the user access to a web-UI to inspect the simulation results.

The AVL SCENIUS™ toolchain enables a companywide collaboration and traceability of all scenario-based tests. The holistic solution covers the entire workflow from requirements engineering, via scenario creation and management, to test case planning, execution, and reporting. Thanks to native support for the ASAM OpenSCENARIO® standard, all modules also work standalone and integrate seamlessly into any existing tool landscape.

Blog ADAS Analyzing Critical Conditions
 
 
Watch the Whole Scenario Analysis Workflow
Video file

Stay tuned

Don't miss the Simulation blog series. Sign up today and stay informed!

More About This Topic

AVL SIMpulse – Simulation for Safety – Next-Level Autonomous Vehicle Development
AVL SIMpulse – Simulation for Safety – Next-Level Autonomous Vehicle Development
Digital

The fifth part of this online event series focuses on our simulation solutions for automated and connected mobility. Join us and benefit from our know-how of efficient and reliable virtual validation. Hear from industry experts, engage in insightful discussions, and be part of shaping the future of autonomous driving.

AVL Scenario Simulator™
Introducing AVL Scenario Simulator™: A Free Simulation Platform for Scenario-Based ADAS/AD Testing
On-demand

The new AVL Scenario Simulator™ is a cost-optimized solution for fully automated, large-scale ADAS/AD testing. Build the optimal virtual prototype for each test purpose.

gl-ast_image-web-webinar-scenius-hands-on_06-24.jpg
Hands-on Workflow Demonstration of AVL SCENIUS™ for Scenario-Based Testing and Safety Argumentation
On-demand

Scenario-based testing is established as an efficient method to test critical traffic situations in a reproducible way. The AVL SCENIUS™ toolchain for scenario-based testing ensures end-to-end consistency and traceability, from the requirements to test reports, enabling a thorough safety argumentation and type approval process.

Scenario Simulator Open Training
Open Training for AVL Scenario Simulator™
On-demand

Our simulation expert will give you a free training where he will walk you through all the typical steps necessary to integrate and test an Advanced Driver Assistance System (ADAS) or Automated Driving (AD) software in a virtual environment.

Like this? Maybe you’ll also enjoy these…

gl-ast_blog-battery-aging-header-07-2024
Gaining Insights Into Battery Aging With the Virtual Twin

The battery is undoubtedly the most complex component of modern electric cars and is largely responsible for the driving experience and range. However, over the course of its service life, it is subject to a continuous loss of performance due to degradation mechanisms that impair its storage capacity and thus the range and power output of the vehicle.

gl_ast_image_blog-header_template_04_23.jpg
Particle-Based Simulation to Optimize Dishwasher Design

Dishwashers are one of those common household appliances that can be found in almost every modern kitchen. Over the decades, not only have dishwasher designs and capabilities been adapted, but the methods and technologies used to analyze and enhance various aspects of dishwasher efficiency have also changed significantly.

gl-ast_blog-fast-charging-header-01-07-2024
Optimizing Fast-Charging Strategies for Electric Vehicles

Electromobility is facing a key challenge: battery charging times must be minimized in order to increase the acceptance of electric vehicles. This is of key importance as, alongside range, charging time is one of the most important factors for user satisfaction.

gl_ast_image_blog-header_template-01_04_23
Develop and Evaluate Solid Oxide Electrolyzer Systems Through Simulation

Global initiatives and actions to reach long-term climate goals are evidently resulting in the development and industrialization of new electrolyzer systems. Significant growth of announced electrolyzer projects is forecasted each year and solid oxide electrolyzers (SOEC) are one of the most promising technologies.

gl-ast_image-web-blog-soiling-header_00_06-24.jpg.
5 Reasons Why PreonLab Is the Ideal Simulation Software for Vehicle Soiling Simulations

Understanding and managing the influence of soiling on a vehicle is a critical aspect of driving safety that needs to be considered while designing vehicles. Soiling or contamination can be caused by the deposition of fluid and solid contaminants on the vehicle surface due to splashing, harsh weather conditions while driving, ...

gl_ast_image_blog-header_calibration-of-fuel-cells-and-electrolyzers
Advanced Automatic Calibration of Fuel Cells and Electrolyzers in AVL FIRE™ M

As the world seeks sustainable energy alternatives, fuel cells and electrolyzers emerge as a promising solution. These electrochemical devices play critical roles in the clean energy landscape. 

gl_ast_image_blog-header_thermal-runaway
Preventing Thermal Runaway: Simulation as a Tool for Improved Battery Cell Safety

Driving range is one of the key sales drivers of battery electric vehicles and to the end customers, every kilometer counts. There are several ways the total efficiency of the vehicle can be increased, such as improving aerodynamics or decreasing vehicle weight, but one of the major contributors is an efficient thermal management system (VTMS).

gl_ast_image_slideshow-release2024r1_keyvisual_16x9.jpg
AVL Simulation Software Release 2024 R1

Discover new features and updates to our simulation solutions.

Skip to main content Toolbar items Administration menu Home Current page Content Structure Translation Reports Configuration Help Close Breadcrumb Back to site  Edit gl_iodp_imag_optimizing_hybrid_powertrain_system_interactions_on_all_testbed_types_07.22.png  Edit Media Toolbar items Prod Go to  Global Nusa.Viher@avl.com Edit Image gl_ast_image_header-blog_vtms-kolaric_04_23.jpg Primary tabs Edit(active tab) Delete Usage Translate Name gl_ast_image_header-blog_vtms-kolaric_04_23.jpg Category  - None - Statu
Leveraging Simulation to Achieve Highly Efficient Vehicle Thermal Management

Driving range is one of the key sales drivers of battery electric vehicles and to the end customers, every kilometer counts. There are several ways the total efficiency of the vehicle can be increased, such as improving aerodynamics or decreasing vehicle weight, but one of the major contributors is an efficient thermal management system (VTMS).

Stay tuned for the Simulation Blog

Don't miss the Simulation blog series. Sign up today and stay informed!

CAPTCHA
By clicking on submit, you give consent to the use of the data you provided to process your request and to receiving communication in connection with your request/registration.
Please click here to view the AVL Privacy Policy.