Autonomous vehicle

Autonomous vehicles (AV), also known as self-driving cars, driverless cars, or robo-cars, are vehicles that are capable of sensing their environment and moving safely with little or no human input.

AVs use a variety of sensors, artificial intelligence, and control systems to operate themselves, including radar, lidar, sonar, GPS, odometry, and inertial measurement units. Another fundamental element of self-driving cars is their ability to communicate with each other along a string (also called a platoon) to maintain safety along the road, regardless of conditions or situations.

Autonomous vehicles can dramatically reduce the frequency of automotive crashes, with the Insurance Institute for Highway Safety (IIHS) estimating that if every vehicle had forward collision and lane departure warning systems, blind spot assist technology and adaptive headlights, almost 1/3 of all crashes and fatalities could be prevented.

Self-driving vehicles have been a goal and challenge for inventors for hundreds of years, dating back to da Vinci’s Self-Propelled Cart in the 1500’s. Automated aircraft and watercraft were instrumental building blocks to autonomous vehicles, with self-propelled torpedoes and aircraft autopilot coming in 1868 and 1933. Ralph Teetor invented cruise control in 1945 after becoming frustrated by the car’s rocking motion when driving with his lawyer. The Stanford Cart, created in 1961 by James Adams amidst the space race, was a remote-controlled lunar rover that ultimately became the world’s first self-driving wheeled vehicle.

Tesla Autopilot, released in 2015, created one of the biggest kickstarts to autonomous vehicle technology. The feature allowed hands-free control highway driving and was released to Tesla owners in the form of a single software update.

Autonomous Vehicles are currently categorized by the Society of Automotive Engineers (SAE) levels of autonomy. There are 6 autonomy levels (level 0-level 5):

• Level 0: The human driver does all the driving.
• Level 1: An advanced driver assistance system (ADAS) on the vehicle can sometimes assist the human driver with either steering or braking/accelerating, but not both simultaneously.
• Level 2: An ADAS on the vehicle can control both steering and braking/accelerating simultaneously under some circumstances. The human driver must continue to pay full attention (“monitor the driving environment”) at all times and perform the rest of the driving task.
• Level 3: AN ADAS can perform all parts of the driving task in some conditions, but the human driver is required to be able to regain control when requested to do so by the ADAS. In the remaining conditions, the human driver executes the necessary tasks.
• Level 4: The ADAS is able to perform all driving tasks independently in certain conditions in which human attention is not required.
• Level 5: The ADAS on the vehicle can do all the driving in all circumstances. The human occupants are just passengers and never need to be involved in driving.

Currently only Levels 1 and 2 vehicles are in mass production, with Honda claiming they will release the first mass-produced Level 3 autonomous vehicle to Japan in March 2021.

A 2020 IDC report predicts that by 2024 more than 50% of all vehicles produced will have some degree of automation (levels 1-5). Lux Research reports that the self-driving vehicle market has an opportunity to reach $87 billion by 2030. Although there are prototypes and training vehicles categorized as Levels 4 and 5, these vehicles are not expected to be mass produced for the road until 2025 or make up a significant portion of the consumer market until 2030.

Level 0 cars currently make up the majority of the automotive market share. A large number of recent model year vehicles are capable of Level 1, meaning they include a single service feature that assists in driving such adaptive cruise control or automated lane-centering.

While Level 1 vehicles only require one assistance feature, Level 2 vehicles must, at a minimum, involve automation of at least two primary functions. These vehicles usually provide a combination of stop-and-go cruise control and automated lane centering. Although these vehicles are often marketed as "self-driving," they still require partial driver control. Level 2 vehicles are expected to capture 92% of the total market share by 2030, with level 3 making up the remainder.

Autonomous vehicles utilize a number of technologies, separately and as a connected system, in order to run efficiently and safely as an individual unit or as part of a "string." Common technologies include telematics and communication systems, computer vision and environment imaging technologies, and navigation and tracking technologies.

Telematics systems utilize cloud-based data transfer to determine vehicle insight, such as speed, location, mileage, and road conditions. These systems have been heavily researched for use in driverless autonomous vehicles due to the importance in programming artificial intelligence driving systems and tracking vehicles without a driver.

Telematics technology applications include driving assistance systems, vehicle monitoring, inter-fleet (or platoon) communication, and emergency response. Location and usage data would not only help individual owners maintain their vehicle, but also assist businesses in tracking fleet maintenance schedules and usage logs.

As driverless cars begin to be produced, telematic services will be important in transitioning vehicles from individually-driven to operation along an interconnected vehicle network.

Radar is an environment detection technology that utilizes radio waves to sense objects, and has been present in cars dating back to the late 1990s. Radar is a cheap and reliable technology that is widespread in vehicles due to its effective at detecting other cars.

Light detection and ranging, lidar for short, is a technology built on laser technology. Lidar shoots millions of laser beams out around a vehicle every second, recording how long they take to bounce off the nearest object and return. An on-board processor analyzes these measurements and uses the data to create a three-dimensional map of the vehicle's surroundings. This map is not only more accurate than traditional radar, but is also easier for computers to process than traditional 2D camera imagery. While lidar is effective for obstacle detection and virtual environment mapping, it is expensive and hard to manufacture at scale.

Autonomous vehicles utilize cloud-based data transfer between moving vehicles to facilitate driver assistance, efficient navigation and reduce traffic congestion. Some AVs utilize map systems, run via cloud and on-board computers to choose routes based on the shortest travel time. Telematic system technology is heavily researched for use in autonomous vehicles, providing key data for monitoring vehicle location, operating on-board systems and communicating with larger networks.

The automotive transportation has evolved in many ways since the automobile was first introduced, both solving and also creating problems in multiple industries since it has grown. Autonomous vehicle fleets have received widespread attention and investment, as large corporations and small businesses try to deal with driver shortages, compensation, and drive-time regulations. The concept has attracted a range of companies to autonomous vehicle research, surging development of fleet management systems, network-to-vehicle communication and coordinated driving technology.