Self-driving cars are a rapidly evolving technology which only a few years ago was still considered science fiction. In such a dynamic context, quick intuitions can be very misleading and misconceptions about the technology, its impact, and the nature of the innovation process abound. In a short article we examine the following four misconceptions:
Oxford’s mobile robotics group has been making rapid progress in the development of driverless cars. As Prof. Paul Newmann explained in a lively lecture last Thursday (as part of the 14th Annual Robotics Systems Conference), it took his group of 20 PhD students just 4 months to build an autonomous car that was able to navigate local streets.
Prototype Autonomous Car (Photo: Hars, 2013)
While being equipped with some algorithms for obstacle detection, the car primarily serves as a test bed for advanced navigation algorithms. Similar to Google, the group uses prior knowledge about the roads to be traveled, but their algorithms can work with much simpler and much less expensive sensors. The car does not need 3D LIDAR sensors. It uses a much cheaper 2D Lidar which is affixed to the very front of the vehicle. The rotating laser captures a slice of points with distance information in a single line below the car as well to the right and the left of the car. As the car moves forward and scans line after line a 3D picture gradually emerges. The car determines its position by comparing the data points gathered to its prior knowledge. The sensor can capture about 40 lines per second. This works well for low speeds but would have to be increased for higher velocities.
Prof. Newmann has also come up with a new approach for navigating in snow and rain. Localization can be very difficult when snow changes the environment’s appearance. His solution is only seemingly simple: instead of trying to detect invariant properties of the landscape, he proposes to accept that the environment may have multiple appearances. Thus he adds the different ways that the environment may look to his store of prior knowledge. As the car drives a known area, it identifies that prior view (winter, summer..) which most closely matches the data captured by its sensors and uses it for localization. It will be interesting to see how robust this approach of “experience-based navigation” can be and how many variations of the environment will be needed to allow fully autonomous driving.
The group currently has two driverless car prototypes; one of them is part of a cooperation with Nissan. It will be interesting to see whether Nissan will incorporate some of the groups navigation algorithms into their solution.
Students at the Ecole Polytechnique of Lausanne may soon drive across campus in up to 6 driverless shuttles developed by French company Induct. The Navia shuttles, of which the first was delivered to the university in December for testing, operate autonomously with a speed of up to 20km per hour. They are fully electric, are equipped with GPS, laser sensors, 3D cameras and can transport up to 8 persons. The shuttles are ideal for last mile transportation. As the laws in most European countries still require all cars to be operated by a driver they currently only can be operated in private areas – such as airports, business and amusement parks, shopping malls, university campuses etc. By removing the first/last mile hurdle, Induct’s shuttle technology has great potential for making public transport more appealing and effective. Compared to individual autonomous vehicles, they are also much easier to justify economically because the high costs of current autonomous technology (especially 3D sensors) are less of an issues for multi-passenger vehicles which clock so many more operating hours than private cars.
Source and copyright: http://www.induct-technology.com
Induct is not the only company focusing on autonomous shuttles. Google operates (or has operated?) a fleet of autonomous golf carts on their campus. Robosoft, another French company also offers 2 types of such shuttles, which have been developed in the European CityMobil research project).
The technology certainly has great potential to become a starting point for more efficient and environment-friendly autonomous people movers and buses. Hopefully the legal framework will be adapted soon to allow the operation of such shuttles in public. This applies especially to European countries which have been heavily financing research in such autonomous transportation systems for almost a decade (and are continuing to do so e.g. in the new CityMobil2 project).
Every year robot trials are conducted in Europe (‘European Land Robot Trials’). The event alternates from year to year between military (M-ELROB) and civilian scenarios (C-ELROB). This years military-oriented event took place Thun, Switzerland from Sept 24 to 28.
14 teams participated performing a variety of autonomous and partiallly autonomous tasks including intelligent surveying and reconaissance, follow-the leader, transport and mine detection.
A German autonomous car prototype based on a Volkswagen Touareg developed by the Universität der Bundeswehr in Munich participated in some of the trials.The MuCAR-3 – shown below has participated in several previous ELROB trials with great success. This time, however, the prototype did not perform quite as well and did not take home any prizes. One reason for slower than usual progress in the past year may have been changes within the research team. It will be interesting to see what the team, which focuses on all-terrain navigation and driving, will be able to showcase next year.
Driverless technology researchers gathered at the beginning of June for the IEEE Intelligent Vehicles Symposium. With almost 200 presentations from more than 600 authors probably no aspect of this technology was left untouched.
This was not just an academic get-together: many of the papers involved major car makers (BMW, Toyota, Daimler, Renault, Volvo, Opel, Volkswagen, General Motors, Hyundai) or automotive suppliers (Delphi, Bosch).
The conference started with a reportedly captivating keynote presentation by Google’s Chris Urmson. Unfortunately, I have not been able to obtain more detailed information about its content. Please contact me if you were there!! Robert Bertini (Intelligent Transportation Systems Lab) gave another keynote on the environmental issues related to intelligent transportation which took the perspective beyond technical issues towards societal and environmental impacts.
It is hard to pick out the most interesting papers. But Daimler presented a new approach for improving stereo vision using a ‘Stixel’-based approach for object recognition. They claim that they are able to reduce false positives by a factor of 8 over the state of the art while reducing the computational costs by a factor of 10.
China also seems to be moving ahead with driverless technology. Two papers (1, 2) were presented from participants of the annual Chinese driverless vehicle competition (‘ Future Challenge of Intelligent Vehicles’) funded by their National Nature Science Foundation.
French research powerhouse INRIA intensifies their research on autonomous vehicles. Having participated in various EU projects (Cybercar, Cybercars2, HaveIT) which looked at intelligent transport systems where coordination between cars played a major role and various aspects of driver-assistance systems, they appear to readjust their focus on individual autonomous cars driving in urban traffic. They are offering a 3-year research position for developing a prototype (and completing a doctoral thesis). This work will include cooperation with business partner Valeo, a one-stop provider of all types of sensors needed for autonomous driving.
The project’s initial goal is to develop an autonomous vehicle for driving at slow speeds in urban settings on private roads around Paris. They will have a lot of work to do to catch up to Google’s self-driving car.