Autonomous Car – An Introduction

An autonomous car is a vehicle that can guide itself without human conduction. This
kind of vehicle has become a concrete reality and may pave the way for future systems
where computers take over the art of driving. An autonomous car is also known as a
driverless car, robot car, self-driving car or autonomous vehicle. Autonomous cars are
commonly classified by self-driven or driverless. Self-driven cars are not fully
autonomous and needs intervention of human drivers. Most of the commercially
available autonomous cars today are self-driven cars. On the other hand, driverless
cars are fully autonomous and can drive without any human interventions.
Autonomous cars use various kinds of technologies. They can be built with GPS
sensing knowledge to help with navigation. They may use sensors and other equipment
to avoid collisions. They also have the ability to use a range of technology known as
augmented reality, where a vehicle displays information to drivers in new and innovative
ways.
Some suggest that significant autonomous car production could cause problems with
existing auto insurance and traffic controls used for human-controlled cars. Significant
research on autonomous vehicles is underway, not only in the U.S., but also in Europe
and other parts of the world. According to some in the industry, it is only a matter of time
before these kinds of advances allow us to outsource our daily commute to a computer,
Potential benefits include reduced costs, increased safety, increased mobility, increased
customer satisfaction and reduced crime. Safety benefits include a reduction in traffic
collisions, resulting in injuries and related costs, including for insurance
Another characteristic of autonomous vehicles is that the core product will have a
greater emphasis on the software and its possibilities, instead of the chassis and its
engine. This is because autonomous vehicles have software systems that drive the
vehicle meaning that updates through reprogramming or editing the software can
enhance the benefits of the owner (e.g. update in better distinguishing blind person vs.
non-blind person so that the vehicle will take extra caution when approaching a blind
person). A characteristic of this reprogrammable part of autonomous vehicles is that the
updates need not only to come from the supplier, because through machine learning
(smart) autonomous vehicles can generate certain updates and install them accordingly
(e.g. new navigation maps or new intersection computer systems).
Problems include safety, possible technological errors, liability, legal
framework and government regulations; risk of loss of privacy and security concerns,
such as hackers or cyber terrorism; concern about the loss of driving-related jobs in the
road transport industry; and risk of increased suburbanization as travel becomes more
convenient
MAJOR COMPANIES AND STARTUPS POWERING AUTONOMOUS
systems

Levels of Autonomy

According to the National Highway Traffic Safety Administration, autonomous cars are
segmented under six different levels of autonomy starting from Level 0 to Level 5. The
following table explains the autonomy levels in the vehicles.
MAJOR SYSTEMS INCORPORATED IN THE AUTOMONOUS SYSTEMS

Level 0: No automation
A car has no automated assistance technologies, though it may feature traditional fixed-
speed cruise control hardware or warn of an impending crash (without intervening). A
vehicle that fits into this category relies on a human to dictate every driving action.
System capability: None. • Driver involvement: The human at the wheel steers, brakes,
accelerates, and negotiates traffic. • Examples: A 1967 Porsche 911, a 2018 Kia Rio.

Level 1: Driver assistance

Most modern passenger cars qualify as capable of Level 1 on the SAE scale. To meet
this requirement, a vehicle must have at least one advanced driver-assistance feature --
adaptive cruise control, for instance. Mobility is still supervised by a human, but for
convenience, the vehicle is capable of maintaining its own speed under certain
circumstances. Lane-keeping tech also falls into this category.
  System capability: Under certain conditions, the car controls either the steering or
the vehicle speed, but not both simultaneously.
 Driver involvement: The driver performs all other aspects of driving and has full
responsibility for monitoring the road and taking over if the assistance system
fails to act appropriately. • Example: Adaptive cruise control.

Level 2: Partial automation

A Level 2 vehicle has two or more advanced driver assistance systems (ADAS) that can
at times control the braking, steering or acceleration of the vehicle. Examples of
qualifying ADAS include adaptive cruise control, active lane-keep assist or automatic
emergency braking, and these technologies must be applied in a coordinated fashion.
Such individual assist features vary in sophistication, but are increasingly common, and
are available on nearly all but the most budget-minded vehicles in 2018. However, it's
the coordination between two or more of these assist technologies that helps them
qualify for Level 2 status.
Importantly, in a Level 2 vehicle, a human driver must still actively monitor the vehicle's
progress and be ready to intervene at any time. System capability: The car can steer,
accelerate, and brake in certain circumstances. Driver involvement: Tactical maneuvers
such as responding to traffic signals or changing lanes largely fall to the driver, as does
scanning for hazards. The driver may have to keep a hand on the wheel as a proxy for
paying attention. • Examples: Audi Traffic Jam Assist, Cadillac Super Cruise, Mercedes-
Benz Driver Assistance Systems, Tesla Autopilot, Volvo Pilot Assist.

Functions that make partial automation possible are already a reality and in practice.
Semi-autonomous driving assistance systems, such as the Steering and Lane Control
Assistant including Traffic Jam Assistant, make daily driving much easier. They can
brake automatically, accelerate and, unlike level 1, take over steering. With the remote-
controlled parking function, it’s possible to pull into tight spots without a driver for the
first time. In level 2, the driver continues to remain in control of the car and must always
pay attention to traffic.
Here’s the list of some driving assistance systems,
1. Adaptive Cruise Control automatically adapts speed to maintain a safe distance
from vehicles in front.
2. Autonomous Emergency Breaking detects an obstacle, warn the driver or
automatically brake to avoid or mitigate a crash.
3. Lane Detection using a forward camera to detect lane markings on the road.
4. Lane Keeping Assist combines a forward-facing camera to detect lane markings
with an electric steering system, keeping the vehicle in the center of the lane.
 5. Parking Assistance systems are designed to help a driver park. Some perform
the entire job automatically, while others simply provide advice so that the driver
knows when to turn the steering wheel and when to stop.
6. Parking Line Detection system that detects markers on the road surface in order
to determine the exact location of parking lots.

Level 3: Conditional automation

The jump in complexity between Levels 2 and 3 is huge compared to the jump between
1 and 2. A Level 3 vehicle is capable of taking full control and operating during select
parts of a journey when certain operating conditions are met.

For example, a vehicle that is capable of managing itself on a freeway journey,

excluding on- and off-ramps and city driving, might be considered Level 3 automated.
This level of automation requires advanced sensor packages, hardware backups and
sophisticated software to keep occupants safe.

The driver must remain vigilant, even when the vehicle is self-driving, in the event of a
failure. Even with Level 3, a driver monitor system is all but a prerequisite to ensure that
the person in the driver's seat is sufficiently alert to take over when conditions dictate.

Google achieved Level 3 autonomy back in 2012 with its test vehicles, but found that
human drivers were too trusting and slow to retake control from the system in the event
of trouble. This observation ultimately led Google to decide against taking the tech to
market, so it's pursuing full Level 5 automation through its Waymo division.

System capability: In the right conditions, the car can manage most aspects of driving,
including monitoring the environment. The system prompts the driver to intervene when
it encounters a scenario it can’t navigate. • Driver involvement: The driver must be
available to take over at any time. • Example: Audi Traffic Jam Pilot.

Level 4: High automation

Level 4 is where things start getting a little "Minority Report," and where bona-fide
autonomous driving systems kick in. A Level 4 vehicle is capable of completing an
entire journey without driver intervention, even operating without a driver at all, but the
vehicle does have some constraints. As an example, a Level 4 vehicle may be confined
to a certain geographical area (i.e. geofenced), or it could be prohibited from operating
beyond a certain speed.

A Level 4 vehicle likely still maintains driver controls like a steering wheel and pedals for
those instances in which a human may be required to assume control.

System capability: The car can operate without human input or oversight but only
under select conditions defined by factors such as road type or geographic area.
• Driver involvement: In a shared car restricted to a defined area, there may not be
any. But in a privately owned Level 4 car, the driver might manage all driving duties on
surface streets then become a passenger as the car enters a highway.
• Example: Google’s now-defunct Firefly pod-car prototype, which had neither pedals
nor a steering wheel and was restricted to a top speed of 25 mph.

Level 5: Full automation

Level 5 is the ultimate goal of self-driving vehicle developers. A Level 5 vehicle is
capable of complete hands-off, driverless operation under all circumstances. This is the
level where there are no provisions for human control -- no steering wheel, no pedals,
no joysticks. A vehicle's passenger would be able to, in theory, kick up their feet
and play some Fortnite, wholly unconcerned about the act of driving.

A Level 5 autonomous vehicle is unconstrained geographically and theoretically able to

travel at all speeds in safety, thanks to advanced software and vehicle-to-vehicle and
vehicle-to-environment communications.

General Motors has been the most vocal about pushing for a commercially viable Level
5 vehicle in the near future, with its Cruise autonomous test vehicles. However, many in
the industry are concerned that GM may have developed an Icarus complex, and is a lot
further from true Level 5 autonomy than it thinks.

It's hard to imagine a world where Level 5 autonomous vehicles become the norm,
available to all. If that happens, how would that change the way that we live? We'd no
longer need to be anywhere near as concerned with traffic or traffic safety. We'd be able
to work during commutes or dispatch vehicles to run errands for us while we did
something else. We might not even need to own a car at all, since they should be easy
to summon from a service.
Ultimately, it will likely be a long time before Level 5 becomes a reality, if indeed it
actually happens at all. But even if it takes decades for a majority of new vehicles to
feature full autonomy, it's exciting to think about right now, and it's important to
understand how the level-by-level ramp-up is affecting motoring life today.

System capability: The driverless car can operate on any road and in any conditions a
human driver could negotiate. • Driver involvement: Entering a destination.
• Example: None yet, but Waymo—formerly Google’s driverless-car project—is now
using a fleet of 600 Chrysler Pacifica hybrids to develop its Level 5 tech for production.

Present market at level 4 cars on road during 2020

According to the top car-makers, autonomous vehicles (Level 4) will start to hit the
roads by 2020. However, most of the research and consulting firms believe that is will
not be that soon. According to the popular industry predictions, level 4 cars will only
gain some market share in 2025, while level 5 cars may be a reality post 2030. Several
complexities are expected to bar the launch of highly autonomous vehicles soon. One of
the major issue will be the risk factor associated with it. Tesla’s autonomous car testing
in 2016 was a disaster and might impact the decision making of the regulatory
authorities on launching of such vehicles.

Another challenge is to mix the existing fleet of vehicle with the new ones. There are
already millions of cars on the road across the world; hence, it will be difficult to manage
the traffic with both human-driven and autonomous cars.

Case: According to California’s Department of Motor Vehicles, there were nearly 2578
failures in the autonomous cars segment in 2016 alone. Alphabet’s spin-out company
Waymo conducted the biggest testing program in 2016 with 635,868 miles of testing
that accounted for over 95% of total miles driven by autonomous vehicles in California.
Waymo’s 60 self-driving cars reported a total of 124 disengagements, 51 of them due to
software problems
Even though developed western countries of North America and Europe have taken
significant steps to encourage AVs on road, Asian countries such as China, Japan and
Singapore is expected to catch up fast. According to analyst estimates, by 2013, China
might become the leading users of semi-autonomous and highly autonomous cars.

Singapore government is equally encouraging test drives of autonomous vehicles. It has

clearly stated its intention to make Singapore a smart nation encouraging AVs that are
efficient, environmentally friendly, and time-saving. Singapore Ministry of Transport also
believes that within next 10-15 years AVs will be hardcore reality.

On the other hand, Japanese officials announced their intention to make Tokyo a self-
driving city before Olympics 2020. In September 2016, Japan's top automotive-tech
houses begun the work of 3-D mapping the country's roadways to get them ready for
autonomous vehicles.
Major potential knowledge required for developing Autonomous cars
1. C or C++ / GPU programming

2. Python

3. Image processing / Video processing

4. Artificial intelligence (AI)

5. Machine learning/ Deep learning

6. Git

7. Matlab / Simulink

8. Java

9. Shell script

10. Embedded software

11. Learn multivariable calculus

12. Learn linear algebra

Algorithm requirements for following hardware systems connected with

camera and sensors using GPU

1. UNSUPERVISED LASER CALIBRATION

2. MAPPING AND LOCALIZATION
3. OBJECT RECOGNITION
4. TRAJECTORY PLANNING
5. DYNAMICAL MODELING AND CONTROL
6. TRAFFIC LIGHT DETECTION
7. GENERIC SIGN DETECTION AND DIRECTION INVARIAT STOP SIGN
CLASSIFICATION
Systems required for a complete autonomous vehicle
Before starting Autonomous vehicle…consider these gap..

As an academic research lab, we have to focus to make efforts on scientifically

interesting challenges with important practical implications. We consider tasks such as
object detection and classification, precision localization and planning under uncertainty,
and automatic calibration and environmental feature discovery, to be among the most
algorithmically demanding topics that were not fully solved by any Urban Challenge
entry or patent.

Nevertheless, much work remains to be done before self driving cars become a reality
for commuters. Significant engineering effort, beyond that appropriate for a research
lab, must go into a system to ensure maximal reliability and safety in all conditions.
Sensors, of which several hundred thousand US dollars’ worth are used on our vehicle,
are still prohibitively expensive for a consumer vehicle. Finally, the hardest perception
and reasoning tasks still remain unsolved to date, as no autonomous vehicle has yet
demonstrated an ability to understand and navigate construction zones, accident areas,
and other unexpected scenarios at nearly the proficiency of a human driver