Our imagination reaches reality. Concept of autonomous vehicles isn’t that new. The idea might have struck everyone of us either in our imaginations or at least from the recently watched sci-fi movies. Well, guess what? it is happening right now. A number of them are currently developing around different parts of the world. Even though the technology has not made it fit for the common man’s use, they are on roads for their test drives.
One of the main aims of this perception model is to eradicate careless driving and make the roads more safe and convenient. The number of road accidents and deaths due to careless driving is skyrocketing at an alarming rate. Autonomous vehicles, on the other hand, are expected to be more smart and vigilant than human. Connected vehicle is another implementing feature of this technology enabling communication between the vehicle and its nearby obstacles. Introduction of these techniques can deliberately diminish the percentage of accidents. According to various sources, many of the developed autonomous vehicles that are into the test drives have successfully completed thousands of miles, themselves proving to be the safest vehicles on the roads in the near future.
Global Race For Autonomous Vehicle
Nowadays implementation of autonomous vehicle has become a global race. Many tech giants and startups are experimenting with autonomous cars. The United States, United Kingdom, Germany, Netherlands, Singapore, Sweden, Japan are some of the leading contestants in the race for autonomous vehicles. When the whole globe is after the technique, where is India in the race? The answer is not too behind. Even though Indian roads are infamous for highly crowded traffic the technology is developinging in the country. However, other features of Indian roads like traffic management, negligence of riders, lack of strict enforcement of speed limits, helmet violations, etc. have made the implementation of these techniques a bit more challenging as compared in other countries. Hence, the training of a vehicle compatible to Indian roads are complex and anarchic as Indian roads itself.
How They Work ?
Coming to the technology part, what makes self-driving cars possible? The self-driving vehicle is now a reality because of the advancement of the hardware and software part. The technology has been developing by many tech giants like Google, Uber, Tesla,..etc.
While looking at the hardware details, most of the autonomous vehicle works on a wide array of sensors, like radar. Uber’s self-driving car use one LiDAR and 360° radar coverage along with other sensors. Google uses eight sensor camera, an array of 32 or 64 lasers to measure the distance. Software process the inputs, drive a path and send instructions to the vehicle actuators which help to control the steering, accelerators and breaking. Algorithms such as obstacle avoidance, predictive modeling and object discrimination help software to navigate obstacles and follow the traffic rules.
Semantically segmented image of a street view
Deep learning is the software advancements that happened and are capable of enabling the highest level of automation(Level 5 discussed in the next section). This software rely more on the previous data from which the next possible solution or decision can be made depending on certain patterns. The previous data can vary from images to LiDAR sensor data as mentioned earlier. Amog them, images are the most commonly using type but can’t be used in its raw form for the purpose of training. In order to make the images/videos fit for training, it has to be labeled by the image annotation technique.
Levels Of Automation
According to the capability of self-driving, researchers have scaled the vehicle in 5 levels from level 0- level 5.
No Automation(Level 0):
It requires a human driver to control. All systems such as steering wheel, brakes, acceleration, and traffic negotiation are under the discretion of the driver. The current version of vehicles on the roads comes under this category.
Driver Assistance(Level 1):
This level include certain systems such as the mechanism to control the steering or vehicle speed, but not both simultaneously. But still, the driver will be responsible for all other aspects of driving. Assisted self-parking is an example. It falls under the most basic level, though there are newer and more advanced features that take it to another level. Adaptive cruise control feature found in some high-end cars is another example. Increasing and decreasing the speed automatically helps the driver to control the speed and ensure safe driving.
Partial Automation(Level 2):
This is the initial level of semi-autonomous cars. The vehicle allows at least two simultaneous automated functions, like braking and steering. However, it does not engage automatically and need a driver to initiate. All the controlling of traffic tactics such as changing lanes and avoiding road hazards are still under the control of humans. Examples of this systems include Tesla Autopilot, Cadillac Super Cruise, Audi Traffic Jam Assist, Volvo Pilot Assist, Mercedes-Benz and Driver Assistance System.
Conditional Automation(Level 3):
The car can manage all actions, but the driver is expect to take over when alerted. There are some road conditions at that time it requires the driver to take over the control. If the car has any issues in the functioning of sensors or unfamiliar conditions driver can take the wheel. In Tesla Autopilot and Audi traffic jam pilot are using this technology.
High Automation(Level 4):
High automation means the car is fully-autonomous in the driving scenarios and can plan a route and arrive at its destination with minimal intervention. The driver has to set the route for the car to navigate. Then, the car will take control and get the passengers to the destination they have routed. The passengers need not mediate in between but can do so when necessary.
Full Automation(Level 5):
The self-driving car is completely capable of self-driving without a human in every situation. It is full autonomy to operate at any road and traffic conditions. It is capable of picking itself and dropping passengers without any human interface. The car not only drives itself but can make its own decisions when interfering with road traffic. Waymo by Google is one of the examples of self-driving cars.
Impact Of Autonomous Vehicles
As this technology became accessible to the public the impacts are sure to be significant. There are many positives and negatives while implementing autonomous vehicles in society. The major goal of the driver less car is safety benefits. Most of the road accidents are happening because of human mistakes. They are expect to reduce traffic accidents. For that software could prove to be less error-prone than humans but cyber security is still a chief concern and so we can say that self-driving car can ensure safety while driving.
Self-driving technology could also be helpful for the individuals who are disabling or the individuals who are not capable to drive themselves. So, by implementing this they will get an opportunity to move without depending anyone. Relieving traffic flow can reduce traveling time. With driverless vehicles, a portion of this commute time can be using for reading, working, relaxing, accessing entertainment and can have a relaxed life.
On the other hand, one of the vehicle cons. energy consumption is an important factor nowadays. The energy consumption is two times more than that of normal vehicles. As an autonomous vehicle is more easy and convenient the number of travel will be increased affecting our environment badly. While discussing the job opportunities, it can replace the drivers negatively affecting public transportation funding.
From this a simple conclusion that can be drawn is, if our society is committed to a safe and efficient driving strategy, self-driving vehicles will run all over the world despite all the limitations. Even though the autonomous vehicle has importance in terms of practice, they are not practical without economic and social consequences.