Margaret Rouse is an award-winning technical writer and teacher known for her ability to explain complex technical subjects simply to a non-technical, business audience. Over…
Navaneeth Kamballur Kottayil is a Senior Machine Learning Developer in AltaML. He has bachelor's degrees in Electronics and Signal Processing, masters (IIT Kharagpur, India) and…
An autonomous vehicle is a vehicle that can drive itself without input from a human driver. There are several types of self-driving vehicles, depending on their level of automation. These levels have been defined by the Society of Automotive Engineers (SAE) which has set 6 of them adopted by the U.S. Department of Transportation ranging from Level 0 (fully manual) to Level 5 (fully autonomous).
Autonomous vehicles rely on advanced artificial intelligence (AI) and machine learning (ML) systems to “understand” their environment and react to commands. Complex sensors and actuators, together with advanced computer vision functions are used to create a constantly updated map of their surroundings, detect the presence of nearby vehicles and pedestrians, measure distances and detect uneven surfaces in roads and sidewalks. Autonomous vehicles can also be interconnected with other external devices such as smart traffic lights and roads, although many of these ideas are still hypothetical.
Autonomous vehicles are also known as self-driving cars, driverless cars, or robotic cars. The term self-driving car is becoming a standard as these technologies continue to mature.
In some ways, the story of the autonomous vehicle is a tale of technological evolution that is not finished yet, a gradual advance of integrated artificial intelligence services and sensor-based driver safety features that may or may not eventually replace human drivers entirely.
Autonomous cars must rely on the presence of sensors situated in various parts of the vehicle to sense their surroundings. Lidar is mostly used for ranging purposes, measure distances, identify lane markings and anomalies such as holes and pits on the road.
Radar sensors are used to track other vehicles, while video cameras can be used to read road signs and detect traffic lights. Advanced AI-powered software will then process all those inputs coming from sensors, and generate instructions to the car’s actuators to plot paths, control steering, braking and acceleration, or dodge obstacles.
The earliest edge of the autonomous driving frontier was the evolution of multiple driver safety features that are now standard in many new vehicles. For example, a lane departure warning system alerts drivers if the vehicle seems to be leaving a particular space on a multilane road. Parking assist features, automated braking, and other features also apply. Each one has its own specialized function and controls a given task.
What these systems have in common is that, although they promote the idea of autonomous vehicle design, they still do not approach what some would call a self-driving car. The human driver is still in control, and needs to be in control, but utilizes the warnings and alerts to make better driving decisions.
These forms of partial automation that still require a human driver pertain to the SAE levels 0 to 2. Only when the automated system is in charge of monitoring the driving environment and human intervention is purely optional then a vehicle could be defined as “autonomous” (levels 3 to 5). As we progress toward autonomous vehicle design, intermediate types of autopilot systems are becoming more common, with Tesla’s one being probably the most prominent.
Fully autonomous (Level 5) vehicles are currently being tested, but we’re still far from making them available to the general public. Although the autonomous vehicle has come a long way, it is not yet a common mode of transportation, and various obstacles to adoption apply.
In particular, some of the technologies required for build a functional self-driving car very expensive, making the final cost of each vehicle prohibitive for the general public. Radar and Lidar work for prototypes, but if mass production is achieved, their signals and frequency might interfere with one another.
Many anomalous conditions such as snow, debris or oil may represent a significant challenge whenever they cover lane markings and dividers. Also, it’s still early to trust AI to be smart enough to make consistent decisions or split-second judgement calls in life-or-death scenarios, such as when a pedestrian suddenly traverses the road after a steep turn.
Techopedia’s editorial policy is centered on delivering thoroughly researched, accurate, and unbiased content. We uphold strict sourcing standards, and each page undergoes diligent review by our team of top technology experts and seasoned editors. This process ensures the integrity, relevance, and value of our content for our readers.
Margaret is an award-winning technical writer and teacher known for her ability to explain complex technical subjects to a non-technical business audience. Over the past twenty years, her IT definitions have been published by Que in an encyclopedia of technology terms and cited in articles by the New York Times, Time Magazine, USA Today, ZDNet, PC Magazine, and Discovery Magazine. She joined Techopedia in 2011. Margaret's idea of a fun day is helping IT and business professionals learn to speak each other’s highly specialized languages.
What is Turnitin AI Checker? The Turnitin AI checker is an advanced tool aimed at maintaining the integrity of school...
Maria WebbTechnology journalist
What is ISO/IEC 42001? ISO/IEC 42001 is an international standard that provides a governance framework for implementing and continually improving...
Margaret RouseTechnology Expert
What are Physical Resource Networks (PRNs)? The definition of Physical Resource Networks (PRNs) is that they are a type of...
Nicole WillingTechnology Journalist
Trending NewsLatest GuidesReviewsTerm of the Day