Driverless car

The driverless car concept embraces an emerging family of highly automated cognitive and control technologies, ultimately aimed at a full "taxi-like" experience for car users, but without a human driver. Together with alternative propulsion, it is seen by some as the main technological advance in car technology by 2020.

The challenge

The challenges can broadly be divided into the technical and the social. The technical problems are the design of the sensors and control systems required to make such a car work. The social challenge is in getting people to trust the car, getting legislators to permit the car onto the public roads, and untangling the legal issues of liability for any mishaps with no person in charge.
However, any solution can be broken down to four sub-systems:

sensors: the car knows where an obstacle is and what is around it;

navigation: how to get to the target location from the present
location;
motion planning: getting through the next few meters, steering, and
avoiding obstacles while also abiding by rules of the
road and avoiding harm to the vehicle and others;

control of the vehicle itself: actuating the system's decisions.
In examining every proposed solution, one should look at the
following questions:

Is this truly a complete system?
Does it drive itself door-to-door?
To what degree is the proposed solution a step towards the complete vision,
or is it just a trick?
Is the car 'autonomous', or would it need changes to the infrastructure?
How feasible (technically, economically, and politically) would it be to
deploy the entire solution?
Can the system allow for and include existing vehicles driven by humans, or
does it need an open field?
How would it cope with unexpected circumstances?






Some have argued that the problem is AI-complete -- that a safe and reliable driverless car would need to use all the skills of an ordinary human being, including commonsense reasoning and affective computing. The concern is that driverless cars will perform worse than human beings in emergency situations that require judgement and the ability to communicate with other drivers and police. For example, how should a driverless car react to a man waving a flare in the middle of the road?

Driver-assistance:

Though these products and projects do not aim explicitly to create a fully autonomous car, they are seen as incremental stepping-stones in that direction. Many of the technologies detailed below will probably serve as components of any future driverless car — meanwhile they are being marketed as gadgets that assist human drivers in one way or another. This approach is slowly trickling into standard cars (e.g. improvements to cruise control).

Driver-assistance mechanisms are of several distinct types, sensorial-informative, actuation-corrective, and systemic.

Sensors
Sensors employed in driverless cars vary from the minimalist ARGO project's monochrome stereoscopy to mobileye's inter-modal (video, infra-red, laser, radar) approach. The minimalist approach imitates the human situation most closely, while the multi-modal approach is "greedy" in the sense that it seeks to obtain as much information as is possible by current technology, even at the occasional cost of one car's detection system interfering with another's.
Mobileye is a well respected company who makes detection systems for cars, which are currently only used for driver assistance, but are eminently suitable for a full-fledged driverless car. The system also detects the objects' motion (direction and speed) and can so calculate relative speeds, and predict collisions.

Japanese infra-red article

some things from the DARPA challenge....

Road-sign recognition

Navigation







The ability to plot a route from where the vehicle is to where the user wants to be has been available for several years. These systems, based on the US military's Global Positioning System are now available as standard car fittings, and use satellite transmissions to ascertain the current location, and an on-board street database to derive a route to the target. The more sophisticated systems also receive radio updates on road blockages, and adapt accordingly.

Motion planning

It is a term used in robotics for the process of detailing a task into atomic robotic motions.
This issue, also known as the "navigation problem", though simple for humans, is one of the most challenging in computer science and robotics. The problem is in creating an algorithm that would be able to find its way around a room with obstacles, perhaps accomplishing some task on the way.

Control of vehicle:
As automotive technology matures, more and more functions of the underlying engine, gearbox etc. are no longer directly controlled by the driver by mechanical means, but rather via a computer, which receives instructions from the driver as inputs and delivers the desired effect by means of electronic throttle control, and other drive-by-wire elements. Therefore, the technology for a computer to control all aspects of a vehicle is well understood.

Work done in simulation:
While developing control systems for real cars is very costly in terms of both time and money, much work can be done in simulations of various complexity. Systems developed using simpler simulators can gradually be transferred to more complex simulators, and in the end to real vehicles. Some approaches that rely on learning requires starting in a simulation to be viable at all, for example evolutionary robotics approaches.

Social issues:
Getting people to trust the car

Getting legislators to permit the car onto the public roads

Untangling the legal issues of liability for any mishaps with no person in
charge.

Despair of progress in the foreseeable future: The UK government seems to see
little progress until 2056. See Silicon Networks article and CNET.co.uk News.

Getting people to give up their freedom to drive wherever they want, whenever
they want without the aide of a computer - though mixed systems with some
human driven and some computer driven cars are possible.

Motivations:

As nearly all car crashes (particularly fatal ones) are caused by human driver error, driverless cars would effectively eliminate nearly all hazards associated with driving as well as driver fatalities and injuries (traveling by car is currently one of the most deadly forms of transportation, with over a million deaths annually worldwide). This would be especially helpful to people that drive to bars and inebriate themselves; the ability for a car to shuttle them home would practically eliminate drunk driving crashes.
Having the equivalent of a personal chauffeur would be a great convenience:

Time spent commuting could be used for work, leisure, or rest.

Parking in difficult areas becomes less of a concern as the car can park
itself away from a busy airport, for example, and come back when called on a
cell-phone.

Taxiing children to school, activities and friends would become solely a
matter of granting permission for the car to handle the child's request.

Allow the visually (and otherwise) impaired to travel independently.

One could sleep overnight during long road trips.

A driverless car would also be a boon to economic efficiency, as cars can be made lighter and more space efficient with the absence of safety technologies rendered redundant with computerized driving. Also the technology would make transportation more efficient and reliable: there may be autonomous or remote-controlled delivery trucks dispatched around the clock to pick up and deliver goods. Moreover, driverless cars would reduce traffic congestion by allowing cars to travel faster and closer together.

Social Costs:

The social costs of this innovation are similar to those of other past technologies: Unemployment, expense and the elimination of the "old way of doing things". See also Luddites.
As with any new labor-saving technology, this would lead to mass layoffs in the driving, cargo, and distribution industries. Taxis would also be automated, effectively eliminating a source of income for the less skilled. A similar if smaller impact is expected in the roadside-catering and other ancillary businesses. However, history shows that any such economic impact on jobs leads to economic benefits elsewhere that create employment, though often not for the exact same people displaced by the new technology.
In order to recoup the development costs, and in order to maximise the profit opportunity that any exciting novelty presents, driverless cars will initially be significantly more expensive than manual cars.
However, the overall technology need not be limited to the operation of vehicles. Once successfully implemented for vehicles, this technology could be used to implement all sorts of routine personal and labor assistants for humans. The concept of "machine" would take on a whole new meaning.
Driving as a personal hobby and sport, and indeed the entire car-oriented sub-culture would be effectively eliminated. However, for those willing to pay for the extra feature, there could be an option to switch between manual and automated driving to make up for that.

Discussion & Future:

Some systems control everything centrally, and in some the vehicle is truly autonomous in the sense that it "thinks" about its own situation in the first person - such a system can integrate with Humans that think in first person.

Conversely. a system that centrally manages everything, though easier to build from a conceptual and engineering point of view, would face horrendous economic barriers because of the costs of converting an entire city or country to the new system at once. In order to be compatible with humans the "first person" point of view is key. This is for three reasons:

1. a distributed scheme in which each component (car) takes care of itself reduces complexity
2. a system that has the concept of first-person operation can understand what a human driver is up to.
3. for the human driver to understand what the driverless car is doing, it needs to operate and "think" in as similar a way to a human as practical (and safe).