
Teleoperation helps when the autonomous vehicle gets stuck
If autonomous vehicles – even at SAE level 4 or even 5 – encounter a problem where they don’t know what to do, safety comes first. Most of the time, this means pulling over to the side of the road and stopping. In order to solve such situations safely and quickly in the future, DLR is researching the conditions for teleoperation of these vehicles. In unclear situations, the autonomous vehicle can then request support through teleoperation from a technical supervisor.
“The teleoperation of autonomous vehicles is a bridge for rapidly bringing existing technologies in the field of automation into use and onto the roads,” describes Dr Michael Oehl from the DLR Institute of Transport Systems Engineering in Braunschweig. The focus of the work is autonomous driving at SAE level 4, where the automation takes over all driving tasks. Passengers do not have to intervene, even if it would still be possible via control elements such as steering wheels.
Autonomous vehicles are not yet ready for the market. However, the legal framework for their use is already in place in many countries. In Germany, for example, a corresponding law came into force in July 2021. It allows autonomous driving according to SAE level 4 on special routes or within defined operational areas, for example on public roads in the city, on motorways or in special areas such as airports and logistics yards. This makes it possible, for example, to transport people or goods with driverless shuttles. New automated mobility services can thus take on an important role in the design of a climate-neutral transport system. For operation and liability, teleoperation or “technical supervision” – as it is formulated in the law – plays an important role.
- German draft law provides for general approval of autonomous vehicles
- US approves rules for self-driving cars without steering wheels
“We are at the beginning of the development with many ideas and concepts to make the teleoperation of vehicles efficient and safe,” describes DLR scientist Oehl. “We are currently concentrating primarily on the interaction between humans and automation as well as on the design of the necessary interface, i.e. the human-machine interface, or HMI for short. For example, we are looking at what a workstation for teleoperation might look like, what demands the job makes on people and how it must be designed from the point of view of work and engineering psychology.”
DLR’s research work is currently focused on the interaction between humans and automated vehicles, as well as on the design of the necessary human-machine interface (HMI). The latter applies in particular to the design of the workplace for teleoperation: what requirements does the job place on the teleoperator, what would such a workplace have to look like from the point of view of work and engineering psychology, explains DLR scientist Oehl.
Equipped with monitors, user interfaces and headsets, teleoperators could in future guide autonomous vehicles through difficult situations. For example, if a delivery vehicle is parked in the second row, a self-driving car will not drive across a solid line on its own decision, given the current state of technology, but will stop. Similar situations can arise in the context of road works, which are not yet stored in the system and are therefore unknown to the automation system.
However, before the DLR researchers tackle the countless conceivable concrete traffic scenarios, they still have to solve numerous fundamental questions: How and with what information is it best to map the vehicles to be supervised? How many vehicles can a single person monitor and supervise at the same time? What live images can the cameras in the highly automated vehicles provide to support the technical supervisor in making decisions? How long does the average assistance request take to process? And how long does it take to get into the vehicle’s particular situation and make a decision?
“Vigilance, for example, is particularly important in such monitoring tasks. In research, we use this term to describe a state of constant alertness,“ describes Oehl. „Even if nothing happens for a long time, you have to be able to react quickly. To prevent distraction, switching off and fatigue, the workplace in teleoperation must therefore be specially designed. “You must not be permanently overtaxed – similar to train drivers or in air traffic control – but also not undertaxed,”. In order to investigate these questions more closely in the context of teleoperation of autonomous vehicles, the team has built such a workplace as a prototype. In this way, the researchers are investigating in an application-oriented way what information teleoperators will need in the future at their workplace in order to be able to support autonomous vehicles quickly and safely in situations that are unclear for the car-bound sensors, computers and software.
https://www.dlr.de/EN/Home/home_node.html
Related articles:
Autonomous trucks to take to the autobahn
Embedded brain reading enables better human-robot interaction
The new generation of “software-defined vehicles”
Attack strategy fools lidar sensors on driverless cars
Bosch acquires map know-how for autonomous driving
Autonomous cars learn to drive with foresight
Image © DLR
