We’ve talked before about this project but this is the first time (as far as I’m aware) that the US media has actually talked about it.

The video is from the Autonomos project in Germany, a team from Free University Berlin looking at autonomous vehicles.

A not sure how to embed this one so : Click here to see it.

Thanks Gabriel S for sending this in.

From a little birdy:

Anthony Levandowski is the current head of the Google Self Driving Car program.

Should be interesting to see what comes of this.

After the last post giving the run down on the TRB conference (apologies for the horrendous formatting – my fault), our anonymous reader has filed another report!

Here it is:

Some more thoughts and opinions from the conference – and I fully accept if others at the conference saw things very differently from me – it would be great if other participants would post on here and balance it out.

One key comment from yesterday that I forgot to mention was that Mercedes will continue with their successful business model approach of implementing new systems one at a time on their continual progression towards full autonomy (whoops – should I have written automation?).

In one of the breakout sessions today on Driver Vehicle Interaction a key research topic identified is the need to work out the whole of issue of how disengaged the driver can become in some of these ‘semi-autonomous modes’ like CACC (co-operative active cruise control; a bit like platooning) and how long would it take to re-engage the driver. Which I guess is seeking reassurance of when and how is it safe to disengage the driver and how long does it take to re-engage him. So this would be driving with periods of ‘hands and feet off’ as the car handles longitudinal and lateral movements.

My preference is that we just jump a few steps and disengage the driver completely; to which I ask, is the Google car intended to be a ‘body out’ solution at some point in the future? It would be really interesting to know whether that is what the public wants. Although when you think of the way some companies have managed to sell vast quantities of products that people never knew that they wanted (well I never knew that I had wanted an iPad until I had read about it and handled it) – then maybe there is a market for this type of vehicle.

The breakout session groups discussed a load of topics around the autonomous vehicle and its implications – seeking to develop ideas for research and in part guide a roadmap that will help the US government to implement this technology in an optimal way. A lot of topics that were discussed would be very familiar to DriverlessCarHQ readers and I know I was able to contribute more for having kept up to date with editorials and comments.

My overall impression was that it was a great workshop and some really useful research will result from it.

GM is working on new Vehicle to Pedestrian technology utilizing Wi-Fi. See here and here, original press release is here. The idea is that the cars would be able to detect people within with a wireless enabled device (such as a smartphone) and avoid running over them. However, for effective communication between both the car and the pedestrian the wireless devices may both need to be running an App to share data.

Big Thanks to Gabriel and Chris for posting the links on our Facebook page.

From the annals of  ”how is it possible that someone only JUST thought of this” comes a fascinating article today from Autopia.

The idea is simple – if a car has a collision, it will then slam on the brakes to avoid it hitting a tree or oncoming car. (What if the brakes make you stop right in the path of an oncoming car? Whoops).

The system has been developed by our good friends over at Continental, who are working on a semi-autonomous system as we speak.

In hopes of preventing these secondary crashes, Continental, a German tire, braking, and stability control developer, has created Post Crash Braking Technology that slows the car automatically after an impact.

PCBT, part of the company’s full ContiGuard safety system, detects a crash from the car’s airbag sensors. It then electronically applies the brakes automatically to slow the car to a stop if the driver can’t operate the pedals due to disorientation or unconsciousness from the impact. The system uses stability control data to keep the vehicle from skidding or locking the brakes.

Autopia has another great article out today which is about a Kinect-style system for the car. Soon, we hope, we will be able to run our car using gestures rather than staring at complicated entertainment systems trying to figure the bloody thing out:

Automotive tech supplier Harman apparently has plans to beat Microsoft to the punch and has gone a step further with a system that recognizes facial and hand gestures. The company’s European division has unveiled a concept vehicle with a system that interprets nods, winks and hand movements to execute in-car commands. “It’s all about reducing distractions in the car,” Hans Roth, director of technology at Harman, said at a recent media event.

Many may not know this but there’s a conference on right now in Irvine, California hosted by the Transportation Research Board about driverless cars.

Brad Templeton has blogged his impressions:

I’m here in Newport beach at the Transportation Research Board’s conference on self-driving vehicles. Today in a pre-session there was discussion of pre-robocar technologies and in particular applications of “managed lanes” and what the might mean for these technologies. Managed lanes are things like HOV/carpool lanes, HOT (carpool+toll), reversible lanes etc. Many people imagine these lanes would be used with pre-robocar technologies like convoys, super-cruise, cooperative ACC, Bus Rapid Transit etc.
As I’ve said before the first rule of robocars is “you don’t change the infrastructure.” First you must make the vehicles operate fully on the existing infrastructure. And people are doing that. But we can also investigate what happens next.

Robocars as many envision them do not thus need dedicated lanes, even though some of the simpler technologies might. Earlier we talked about electrification which is a pretty expensive adaptation. Let’s talk about high speed lanes.

He goes on to discuss the idea of dedicated lanes during off-peak times.

I would give good money to be there but am on the other side of the world, unfortunately. However, there’s good news…

We have also had an anonymous reader send in a full run-down (slightly redacted):

Richard Bishop is here – and so many others that are probably familiar names to the DriverlessCarHQ aficionados.

There has been some really interesting discussion already – and one or two subtle digs around the turf war of the connected vehicle piece and the autonomous vehicles piece.

Levandowski’s presentation was ruined a bit by a Microsoft Powerpoint malfunction – very few of the presenters were able to present their embedded videos without some sort of hiccup. Broggi’s presentation was great as they gave gone down the vision route and he basically used his presentation to have a dig at the expensive sensors that the others use as they produce 3-D from their stereo cameras. Later on Luca Delgrossi had a similar dig as Mercedes use stereo vision, but they go one (or more) better and get 6-D! Do these guys know that Ibeo are going to produce the liar sensors at $250 each in a couple of years?…….

Early on today it was made very clear to delegates that we should no longer be talking about autonomous vehicle, but automated vehicles – I think that was in Steven Shladover’s Lexicon and Taxonomy intro . After which all of the speakers being developers of ‘automated’ technology were bending over backwards to apologise for using autonomous in their slides – including most of Levandowski (Google), Chris Gerdes (Stanford), Alberto Broggi (VIsLab), Chris Borroni–Bird (GM) and Luca Delgrossi (Mercedes).

Bryant Walker-Smith made a great presentation on legal aspects, and near the end, on a slide showing that Nevada had already passed laws, added something like “Autonomous – Sorry Steve, but it’s the law” – which earned a roar of laughter from the participants.

Some questions were asked after the talks – Levandowski was asked when their technology would go on sale to the public – I think he answered something like “When it’s certified safe”. The next question was “When is safe ‘safe enough’” and that led to some interesting discussion between the panel of presenters on the subject – hopefully that got caught on the webinar.

There were three presentations on platooning – a lot of commonality between the projects, but each did something different that added to the overall knowledge. The Volvo guy made it clear that their cars took a lot of front end and windshield damage from gravel thrown up from behind the lead truck – and it affected the second and third car too. Not the sort of stuff that leaves me wanting to save a few dollars in fuel.

Its workshop day tomorrow when we break out into discussion groups and start to sort out the areas of research needs and build the data set of questions and comments needed to start a roadmap to take this whole piece forward.

All in all a great workshop and plenty more to come tomorrow.

An article over at Tomsguide.com has a list of Google’s driverless car patents, both pending and accepted. The article itself is brief and worth reading as it discusses the various authors of each patent.

If you are interested I’ve listed links and abstracts below:
Traffic Signal Mapping and Detection

A system and method provides maps identifying the 3D location of traffic lights. The position, location, and orientation of a traffic light may be automatically extrapolated from two or more images. The maps may then be used to assist robotic vehicles or human drivers to identify the location and status of a traffic signal.

Zone Driving

A roadgraph may include a graph network of information such as roads, lanes, intersections, and the connections between these features. The roadgraph may also include one or more zones associated with particular rules. The zones may include locations where driving is typically challenging such as merges, construction zones, or other obstacles. In one example, the rules may require an autonomous vehicle to alert a driver that the vehicle is approaching a zone. The vehicle may thus require a driver to take control of steering, acceleration, deceleration, etc. In another example, the zones may be designated by a driver and may be broadcast to other nearby vehicles, for example using a radio link or other network such that other vehicles may be able to observer the same rule at the same location or at least notify the other vehicle’s drivers that another driver felt the location was unsafe for autonomous driving.

Transitioning a Mixed-Mode Vehicle to Autonomous Mode 

Disclosed are methods and devices for transitioning a mixed-mode autonomous vehicle from a human driven mode to an autonomously driven mode. Transitioning may include stopping a vehicle on a predefined landing strip and detecting a reference indicator. Based on the reference indicator, the vehicle may be able to know its exact position. Additionally, the vehicle may use the reference indictor to obtain an autonomous vehicle instruction via a URL. After the vehicle knows its precise location and has an autonomous vehicle instruction, it can operate in autonomous mode.

Diagnosis and Repair for Autonomous Vehicles

A system and method of controlling a vehicle is provided. In one aspect, the system and method determines the amount of wear on a component of the vehicle and, based on the amount of wear and information derived from the environment surrounding the vehicle (e.g., another vehicle in the path of the vehicle or a requirement to stop at a particular location), maneuvers the vehicle to mitigate further wear on the component.

System and Method for Predicting Behaviour of Detected Objects

Aspects of the invention relate generally to autonomous vehicles. Specifically, the features described may be used alone or in combination in order to improve the safety, use, driver experience, and performance of these vehicles.

If you know of any more patents from Google or another company please let us know and we will add it to this list or put it up in another post.

US traffic deaths have risen 13.5% in the first quarter of 2012. (Thanks to reader Gabriel for the link). The original NHTSA report is here.

Reports like this are a sobering reminder of why most of us are so enthusiastic about self driving cars – the potential to save over a million lives a year from needless traffic deaths.

We get a lot of links posted or emailed to us that just aren’t quite relevant enough to be individual posts. In the spirit of that, here’s our first Saturday Link Fest.

Google has created a ‘neural network’ of over 16bn connections. Its first effort? It taught itself how to recognise a cat here. Thanks Paul Godsmark.

Thanks Gabriel S for the next few:

P2P Car Rental now available through GM’s OnStar here.

P2P Taxi Services now operating in San Francisco : Cyrus Faviar takes a fascinating ride here.

Automate or Perish here. All about the most recent developments in automation of business.

Quick thoughts…

If a truck loaded up with valuable goods is travelling without any humans to stop it, what’s going to stop people from taking that cargo?

Surveillance doesn’t help because balaclavas are a simple counter-measure.

Heavily fortified trucks could impede the road pirates while distress signals are sent.

Electrification – the internal shell of the trucks could be heavily electrified, making it highly difficult for the pirates to get to the goods.

Convoys – A human could potentially travel with massive convoys of, say, 50 vehicles as a protective measure.

Tear Gas/Pepper Spray could be deployed if threats are sensed.

Self Destruct could even be included as a final counter-measure to make pirate raids futile.

Of all the above solutions, I think a heavy lockdown in combination with guarded convoys would be the solution that is reached. What do you think?

For smaller vehicles performing part of a final-mile solution, there would need to still be quite strong protection but without humans required.