Welcome to the Knowledge Agenda on Automatic Driving, an initiative of the Ministry of Infrastructure and Water Management, Department of Transport and the RDW-Vehicle-approval, to provide an online overview of available and required knowledge in the field of automatic driving.
The overview is divided into a number of knowledge domains to map the various facets. In the library you will find an extensive collection of reports, papers and presentations, including summaries and background information. The library is used worldwide. The last report on Ethics was requested 674 times in a short time! About 30 pieces are downloaded every day.
The collection of knowledge documents is managed in Dropbox. With Dropbox you can search directly in the folders with documents and full text. Contact firstname.lastname@example.org to gain access to the Dropbox.
Since 2015 we keep a list of knowledge questions (the required knowledge). Our collection of documents provides an answer to these knowledge questions. New questions are coming up because we are getting further and further into the implementation of “Connected Automated Driving”. The set of knowledge questions includes the topics automated driving and Smart Mobility (ITS). Additional overviews with projects are available here on the ITS theme. Experts on themes also develop knowledge and standards in the Netherlands/EU; an overview can be found here.
The popular knowledge questions are:
Adaptations to the infrastructure required for automated and nonautomated mixing? Categorize road network for automatic driving?
|Dutch: Rapport Zelfrijdende auto’s, verkenning van implicaties op het ontwerp van wegen
Translated: Chapter 3.3
-Adjustments of the cross section and the berm direction (restraint strip, obstacle-free zone, emergency strips) can not be adjusted yet.
Arc radians cannot be adjusted. However, it could be considered that in multi-lane arches the manually controlled vehicles are only allowed in the outer strip / strips. The inner strip / strips can be reserved for automatic vehicles that can optimize their speed on the infrastructure features and preferences of the occupants. There is a risk that manually controlled vehicles will imitate the behavior of automotive vehicles, which could lead to overturning them too fast. The short driving times of automotive vehicles could also be mimicked by human drivers. This also applies to other road sections.
-Overall, the mix of different vehicle types can initially give some unpredictable traffic image. This in particular affects the dimensioning of exchange points. (traffic entering an exiting, weave, intersection, roundabout). The mix of vehicles of different SAE levels can provide an interaction between the different types of vehicles that go against the intuition of human drivers. Automated vehicles behave differently than drivers of non-automated vehicles, or drivers of vehicles with lower SAE levels. Exchange in some places may be too complex for automated vehicles, who are not yet communicating and need more time, due to the greater safety margins than those held by human drivers and braking early. This may lead to a broader dimensioning of exchange points (entry and exit strips, weave). This is consistent with the practice observation that ACC in its current form leads to greater follow-up distances.
For underlying roads, especially the interaction with slow traffic (cyclists and pedestrians) causes many dilemmas. As a result, the situation becomes much more complex, and no (safe) solutions are available at this time. On area-bound roads with separate traffic flows, this also applies (at intersections and roundabouts).
-When traffic is mixed crossroad design and priority rules must still be retained as usual. The automated vehicle has to adapt itself to the human drivers so that confusion is avoided with drivers of non-automatic vehicles. The automated vehicle may need more time to estimate the situation at a crossroads, if no vehicles are available to communicate with in the neighborhood
– Visibility on a roundabout is not a problem for automated vehicles if it can communicate with other traffic, this causes a problem for human drivers though
– (truck) platooning still brings the necessary questions if the aim is to be able to platoon almost continuously (so no splitting at every node) to achieve the benefits of platooning. From the above, the impression is that the possible road design implications in the mixed traffic situation are likely to be rather limited (either you can not change road design as long as there is mixed traffic). In the case of mixed traffic, no change in design are possible, based on what human drivers need to drive safely, smoothly and comfortably. Only on high lane /capacity roads it could be considered to reserve a part of these lanes to automated vehicles, and to redesign this road section (separation in the cross profile of automated and non-automated).
Smart Infra, Eerste schetsonderzoek naar level 4 snelwegen en kruispunten voor zelfrijdende auto’s
“Another deployment scenario for automated driving involves implementing transportation paradigms that provide slow-moving passenger vehicles, for example in urban areas. Consumers could summon such vehicles using a smartphone app and ride them over relatively short distances (see the use case “vehicle on demand”). In particular, companies from unrelated sectors can use image processing, object recognition, and route planning systems—which are already in widespread modular use—to implement transportation models with higher-order automation within a limited geographical range. The arrangements often proposed for market introduction are slow-moving and limited-area vehicles intended to serve what is known as the “first or last mile,” complementary to private automobiles or public transportation. To name one concrete example, these types of solutions could be used to reach bus and urban rail networks in areas where a regular schedule is not feasible due to inadequate infrastructure or financial limitations. Found on (p.199-200): https://www.dropbox.com/s/91n2z7i19wfgzu6/Beiker%2C%20S.%20%282016%29%20Deployment%20Scenarios%20for%20Vehicles%20with%20Higher-Order%20Automation.pdf?dl=0nn Note Joop: In the Netherlands there are tests concerning a connection with the station Ede and the Campus of Wageningen (WEPODS). In 2018 there will be tests with the WEPOD to the airport Weeze in Germany. There are also some test in the region MRDH near Airport Rotterdam.
Indirect answer:nn“There have been numerous predicted monetary advantages linked to the adaptation of Autonomous Vehicles for daily commuting and travelling. First estimations state that self-driving cars can contribute up to $1.3 trillion in annual savings to the United States economy alone, and an expected sum of $5.6 trillion for global savings (Bartl 2015). “nnFound on (p.30): Impact of Autonomous Vehicles on Urban Mobility nn
“Automated driving, with its minimal space requirements and rather equal speed levels, could at least double the existing average road infrastructure capacity. “nnFound on (p.380): Autonomous Vehicles and Autonomous Driving in Freight Transportnn
“Cars driving from work to home can be designed smaller. Since 80% of these rides hold only one passenger. Lanes that are for AV use only can change their driving direction dynamic, to improve road capacity inbound or outbound traffic. Found in Dutch on (p.54): Verkenning technologische innovaties in de leefomgeving
Our annual knowledge report reports on this. It indicates to which knowledge questions answers and research have become available. In December, we will put the subjects and knowledge questions for research and trials into the coming year. Currently, the priorities in the list of knowledge questions (AR + C-ITS) are being worked on by, among others, IenW, RWS, Knowledge Institutions and Provinces, Cities, regions and pilot projects.
On this site you will also find an overview of relevant conferences and events and a collection of films and webinars via the menu. News and current developments are maintained by us through the library and twitter feed (#KARNL). Every week, a lot of knowledge and material is added to the collection, in all knowledge areas.