Why we won’t have autonomous cars without simulation
Share Article
Read More
Sponsor Content
Jan 8, 2024
Invest Ottawa

Simulated virtual environments efficiently emulate physics, weather, on road obstacles and other conditions, all indistinguishable from the real thing

Area X.O, operated by Invest Ottawa, is a research and development complex that helps accelerate time-to-market and commercial adoption of next-gen smart mobility technologies. Image: Area X.O

Simulated virtual environments like Area X.O efficiently emulate physics, weather, on road obstacles and other conditions, all indistinguishable from the real thing

Invest Ottawa logo

This article is Sponsor Content presented by Invest Ottawa

Simulators were once the exclusive domain of airline pilot training. But, today, simulators are used for many purposes including development of Advanced Driving Assistance Systems (ADAS) and connected and autonomous cars.

Simulation provides engineers with the power to see how ideas perform against millions of variables, unlocking the future of mobility. 

To understand how it works, here’s a simple analogy: the simulator lets self-driving software “play a game” that’s sufficiently close to reality. The more it plays, the better it gets.

And, just like the simulated world of the Matrix, your autonomous technology under test can’t tell the difference between virtual reality and the real thing. 

From Physical to Virtual

Area X.O, operated by Invest Ottawa, is a research and development complex that helps accelerate time-to-market and commercial adoption of next-gen smart mobility technologies.  

As a Regional Technology Development Site (RTDS) and a member of the Ontario Vehicle Innovation Network (OVIN), Area X.O has years of helping some of Canada’s top technology talent test their connected and autonomous technology at their smart mobility facility in Ottawa. To be proactive and to help more companies improve and validate their innovations, Area X.O created a simulation portal in 2023.  

This simulation model, powered by Ansys, the global leader in engineering simulation software, ensures companies can reduce time, cost, effort and risk when getting their breakthrough innovations to market faster. This means companies are further ahead in their technology development once they get to physical testing. Currently, simulation projects can access the following Ansys software options: 

  • AVX: autonomous car and sensors; 
  • SPEOS: camera, HMI and optics; 
  • STK: satellite and drone; and 
  • EDT: electronic desktop or antennas, thermal or anything electronic. 

Funded in part by the Government of Canada through the Federal Economic Development Agency for Southern Ontario (FedDev Ontario), the portal provides registered guests free access to virtually interact with Area X.O assets. These include a digital twin of the real gated Area X.O facility, smart intersections, autonomous vehicles, mobile mannequin test targets and a mobile command centre.   

Recently launched, the simulation assets also feature a digital twin of the DARTT (Drone and Advanced Robotics Testing and Training) Zone at Area X.O. Developed in partnership with InDro Robotics, headquartered in Victoria BC with offices in Ottawa, the DARTT Zone helps technology companies test and evaluate the capabilities of aerial and ground robotic technologies.

DARRT Zone simulation still image
DARTT (Drone and Advanced Robotics Testing and Training) Zone. Image: Area X.O

Simulations aren’t playing around

Just how close to reality are simulation models? Here’s an example. 

While a game tries to be realistic enough to fool a human, simulation for a connected and autonomous vehicle (CAV) doesn’t. This simplifies the detail needed for creating actual virtual worlds for AI to explore. 

Another difference between games and CAV simulations is that unlike video games, simulation software focuses on analyzing, organizing and developing safety requirements, making it possible to conclusively determine functional safety attributes without overlooking potential trouble-spots or failure modes. 

Still image of car and mannequin simulation
Simulation of autonomous driving car interacting with a test mannequin. Image: Area X.O

Eight billion miles to safety

Testing requires the ability to run a huge amount of drive miles on any autonomous software before it becomes safe enough. It’s estimated that eight billion miles of driving experience is required to make a machine learning model safer than a human driver.  

The only way to practically and safely achieve that many miles is by creating simulations with enough fidelity that a machine-learning algorithm can gain “virtual miles” that are indistinguishable from the real thing. Take Waymo: in 2020 they had tested their vehicles on over 20 million miles of actual driving, but 15 billion miles of simulated miles. The simulator gave them 750 times more coverage. 

Another reason why simulation miles are more efficient, is the nature of the miles tested. The same stretch of simulated road can more easily be tested under a host of different conditions – day, night, dusk, rain, snow, haze and more.

An autonomous car system needs to properly react to unusual or unsafe conditions that may be extremely rare or may never be encountered within test-driven miles. Whether it’s semi rollovers, bridge failures, or run-away baby strollers, simulation allows repeated tests and refinement of the self-driving system to make sure that it reacts appropriately and quickly to these rare, but potentially fatal situations.  

Using simulation, engineers can feel confident that the algorithms will work properly in each situation.

ANSYS weather conditions still image
Repeating driving tasks under multiple lighting conditions. Image: ANSYS

Virtually testing hardware

Simulation isn’t just dealing with visual images. Any simulation model also needs to map sensors like lidar, thermal and radar with appropriate point-cloud simulations that are adapted to the virtual environment to mimic those sensors.  

Thus, simulation also provides the ability to create software-only analogs to trial new hardware such as lidar technologies or sensor fusion techniques. 

ANSYS sensor simulation still image
Simulation of sensors. Image: ANSYS

Ready to fly 

You might be asking yourself — does it really make sense to trust our safety to piloting techniques learned in a virtual environment? 

Of course, it does; we’ve been doing it for decades. Many sectors and companies create and use simulation products to test their components, systems and system-of-systems.

Pilots are extensively trained on flight simulators as part of their certifications before they are released into the air. And, while planes don’t yet completely fly themselves, good portions of their flight are automated. These are increasingly being tested in simulated environments. 

We’ve already learned to pilot with simulations. Now it’s the car’s turn.  

Could your company benefit from a simulation project to reduce time, cost, effort, and risk to get your innovations to market faster?

All premium simulation projects, which include access to Area X.O simulation engineers and the opportunity to build customized simulation solutions, are currently free until March 31, 2024.  

Sign up here to access the Area X.O Simulation Discovery Portal powered by Ansys or reach out with questions at [email protected].

View Comments (0)
You May Also Like