Vehicle manufacturers, suppliers, knowledge institutes and governments are taking considerable steps to ensure that self-driving vehicles will become a reality on our roads. By developing technology and a testing and validation methodology TNO is helping to incorporate self-driving cars and trucks safely into existing traffic.
“In recent months, there has been a media hype about automated driving,” says Maurice Kwakkernaat, Programme Manager for Automated Driving at TNO, “with reports of new functions in vehicles that are already similar to high levels of automation. Many people started to think that, for example, a Tesla really drives itself without the need of a driver. Reports of accidents and incidents soon followed. As a result, there is now a growing awareness that self-driving cars work just a little differently. As a driver, we still have an important role to play. We still need to be very careful.”
Billions of test kilometres
“This growing awareness also triggers a strong demand for knowledge and research on the topic of automated vehicles,” Kwakkernaat continues. “In the US, for example, tests and pilots are being carried out with self-driving vehicles that reveal that the technology seems to be working well. But we just don’t know yet exactly how safe they really are. In actual traffic.”
“Before a car is approved, it has already driven hundreds of thousands of kilometres”
Traditionally, vehicle manufacturers ensure that new vehicles have undergone necessary road miles. “Before a car is approved by the certification authorities, it has already driven hundreds of thousands of kilometres. After all, you have to let it experience the most important circumstances it might encounter. If you want to do that in the same way with self-driving cars, then you’re talking about putting in billions of test kilometres. That is an impossible task. TNO has developed a scenario based testing and validation methodology called Streetwise based on its extensive knowledge on vehicle technology, safety and traffic behaviour analysis.”
Inclusion in traffic policy
Within the Streetwise programme TNO is realising an effective approach to test and validate self-driving vehicles and to get them safely on the road, together with national and international parties, including Itility, the Dutch vehicle certification authority RDW and Nanyang Technological University. Kwakkernaat: “We are developing a method to optimally integrate self-driving vehicles into the current traffic picture. To do this, it is first of all important to understand and quantify what is happening on the road. We use measurement vehicles that drive on the road collecting as much information as possible about traffic situations. We then summarise all this in parameterised scenarios.”
One such scenario is a car that changes lanes, overtakes another vehicle and returns to its own lane. “There are a lot of ways to execute such a scenario, and we describe these in terms of parameters, such as the acceleration or deceleration of the vehicle. This leaves us with a clear number of scenarios, with a large collection of parameters that describes how traffic behaves. What is left are situations that you don't encounter very often but that can happen, such as a child suddenly crossing the road. So we'll add those scenarios as well.
“Ultimately, this will lead to systems that are better adapted to the behaviour of other traffic”
“The scenarios and parameters are used to simulate and test vehicle systems during development. This enables virtual testing. The industry is looking for this, and governments and road authorities also want to know how vehicles behave in relevant situations amongst other traffic. Like approaching the tail of a traffic jam. This means that systems can be better adapted to the behaviour of other traffic. Because that is where we want to go: better and safer behaviour.”
Platooning with different makes of vehicle
For the logistics sector, too, there is much to be gained from automated driving. In this context, the European multi-year project ENSEMBLE, coordinated by TNO, was recently launched. Kwakkernaat: “With 20 partners, amongst them DAF, Iveco, MAN, Mercedes-Benz, Scania Volvo and the CLEPA interest group, we are going to ensure that different makes and types of trucks can safely platoon together on public roads, amongst other traffic.”
“We are going to ensure that different brands and types of trucks can safely platoon together on public roads, amongst other traffic”
The technology for truck platooning is already available, but each manufacturer has developed its own variant. “This will only work if you can drive with different manufacturer brands in one platoon. That is why, in this project, we are bringing together the suppliers of the truck manufacturers to make agreements at a higher, tactical level of automation that will lead to a safe and interoperable platooning system.”
A big challenge is how to handle the sharing of vehicle information between competing manufacturers and suppliers. “We are going to build a neutral system that brings together the automation technology of the various trucks. Such a system, which works with communication and the exchange of data, is new. So that is where our independent position comes in handy. A large multi-brand truck platooning demonstration is expected in 2021, demonstrating that it can be done safely on public roads.”
More information can be found on the page 'TNO and the self-driving car'. TNO outlines the potential of the self-driving car. Different facets are discussed: technology development and validation, effects on traffic flow, ICT and the human factor. You will also find relevant links to pages on TNO.nl.
Download our paper on StreetWise
Scenario-based safety validation of connected and automated driving.