Daimler introduces robot truck concept
The Future Truck 2025 is a near-series study based on existing semitrailer model Actros, developed within Daimler’s Shaping Future Transportation initiative. The vehicle is controlled by a system Daimler calls "Highway Pilot". This system has a broad range of sensors and computing resources at its disposal. Among others, it is equipped with a radar sensor in the lower area of the front end which scans the road ahead at long and short range. The front radar has a range of range of 250 m and scans an 18-degree segment. The short-range sensor has a range of 70 m and scans a 130-degree segment. The radar sensor is the basis for the Proximity Control Assist and Emergency Braking Assist already available today. The area ahead of the truck is also scanned by a stereo camera located above the dash support behind the windscreen. This is currently the location of a mono-camera if the optional Lane Keeping Assist is ordered. The range of the stereo camera is 100 m, and it scans an area of 45 degrees horizontally and 27 degrees vertically.
The stereo camera of the Mercedes-Benz Future Truck 2025 identifies single- and double-lanes, pedestrians, moving and stationary objects, all objects within the monitored area and also the condition of the road surface. The camera recognises everything that contrasts with the background, and is therefore also able to measure clearances precisely. The front stereo camera also registers the information on traffic signs.
In addition to object and distance recognition, the stereo camera recognises lane markings as a major function for autonomous track guidance.
The road surface to the left and right of the truck is monitored by radar sensors installed in the sides. They are located on the left and right, ahead of the tractor unit’s rear axle. The sensors have a range of 60 m and cover an angle of 170 degrees.
With multiple networked sensors and multisensory fusion algorithms running on several computers in the truck, a complete image of the surroundings is generated. Fusion of the data from the front radar sensor, side radar sensors and front camera by a high-performance multi-core processor in the central computer provides a continuous view of the entire area in front of and beside the truck. For comparison, the human eye has a 150-degree angle of vision, but its focal area is merely a fraction of this.
The sensor system of the Mercedes-Benz Future Truck 2025 already represents the next generation of this technology. The sensors do not only recognise the road edge by the marker lines, but can even identify the course of the road surface by the roadside features (e.g. guard rails or vegetation).
The sensor and camera technology is active throughout the speed range from standstill to the legally permitted maximum truck speed of 80 kmph. By intervening in the steering, it automatically keeps the truck in the centre of its lane. The system also includes a three-dimensional digital map, which is already used for the assistance system Predictive Powertrain Control (PPC). This means that the truck is always fully aware of the road’s course and topography.
In addition, the concept vehicle is equipped for V2V and V2I networking. Every vehicle equipped with this in the near future will transmit continuous information to its surroundings, the CAM (Corporate Awareness Message). The vehicle uses this to announce its presence. The information content includes vehicle position and model, dimensions, direction of travel and speed, any acceleration and braking maneuvers and the bend radii negotiated.
The frequency of information transfer depends on the vehicle speed and the intensity of any changes in its movement. It varies between one message per second when cruising to ten times that interval when changes are significant.
Transmission is via WLAN technology, using the standard, Europe-wide G5 frequency of 5.9 GHz. The basis is the ITS Vehicle Station (Intelligent Transport Systems and Services) on board the vehicle. Communication between vehicles is also standardised following an agreement between a consortium of automobile manufacturers, suppliers, public organisations and research institutions.
With its massive utilization of V2V (aka Car-to-Car or C2C) communication schemes, the Future Truck 2025 is not on the road in isolation, but constantly communicates with its environment. The result is real-time communication between the networked vehicles creating a traffic situation image that cannot be matched by even the most precise radio traffic reports.
In this way information about slow-moving traffic is passed between vehicles in advance, also data on tailbacks and their length and duration, or on roadwork – the data are available to all road users. As the networked vehicles respond automatically, a steady traffic flow and efficient use of the limited infrastructure are ensured. In the event of major problems, early information is provided about automatically initiated route changes to the destination or recommended diversions. In combination with autonomous driving, road traffic will develop into a self-learning system.
The average transport speed will be increased by the improved traffic flow alone, without raising the speed limit, and at the same time the smoother flow will save fuel. This benefits all parties involved in the goods transport sector: the transport operators and their drivers, dispatchers and customers.
Fig 1: V2X communications, a crucial element in Daimler’s autonomous truck vision, help generating a detailed image of the truck’s surroundings. For full resolution click here.
After joining the motorway, the driver of the concept car merges with the traffic flow in the appropriate lane. The truck reaches the preset speed of 80 km/h. The system then prompts the driver to activate the "Highway Pilot". The driver activates it, and the vehicle switches to autonomous mode. The driver receives an acknowledgement reading "Highway Pilot active".
The Future Truck 2025 is on the road independently according to the traffic situation, as no vehicle travelling ahead is needed as a reference to guide it through the traffic – it literally acts autonomously. The Mercedes-Benz Future Truck 2025 operates independently of other road users thanks to networking, not by being daisy-chained with a lead vehicle as was the case in earlier trials conducted within the Sartre project.
If there is another vehicle travelling ahead, the truck automatically adapts to its speed and maintains a set safety distance. It is therefore always possible for other vehicles, e.g. a car cutting in ahead when moving from the overtaking lane before an exit, to do so safely. Here too the safety distance is always maintained.
Though the driver in this operational mode can relax and devote himself to other tasks, it is always possible to resume manual control. Overtaking maneuvers are not part of the concept; if the driver feels inclined to pass a slower vehicle he must switch to manual mode. The same applies to leaving the Autobahn or changing lanes where the road forks.
By automating goods transport, Daimler claims it would be possible to reduce fuel consumption and pollutant emissions. It would also cause the road traffic to become more fluent with less stalls – an aspect particularly relevant against predictions that the goods transport on roads is expected to increase by 50% over the next 20 years.
To make automated goods traffic a reality, an intelligent infrastructure needs to be created, Daimler points out. Inside the vehicles, many building blocks for automated driving are already in place – things like Proximity Control Assist, Stop-and-Go Assist, Active Brake Assist 3, Lane Keeping Assist and three-dimensional maps for the Predictive Powertrain Control as well as telematics platforms and services. Other elements have been developed over the past years but are not yet applied to road traffic, says professor Sabina Jeschke of the RWTH university in Aachen. "Together with a number of breakthroughs such as the IBM computer Watson or the Google Car in the last two years, we are entering a new era in artificial intelligence – namely mass networking. … The "Internet of Things" plays an important role in this. It represents an expansion of the internet: the participants are no longer just people, but also things – such as the sensor system in a car, climatic data stations, process data systems in production engineering, and other systems that directly interact with the environment. Completely new forms of cooperation between technical systems are made possible on this basis – and especially when it comes to road traffic and mobility."
It is generally expected that concepts for automating road transportation will be the dominating topic at the IAA Nutzfahrzeuge commercial truck exposition, which will take place in September in Hannover, Germany.
Related links and articles: