Artificial Intelligence Techniques Used In Autonomous Cars - Research Paper Example

Artificial Intelligence Techniques Used In Autonomous Cars Artificial intelligence (AI) has emerged to be the future in the field of information technology. This paradigm of the science is proposed to develop artificial intelligence into machines and thus automating their actions. In addition, an important mission of artificial intelligence paradigm is to create human like intelligence in machine to reinstate and assist humans in everyday tasks. The basic purpose of this paper is to present a detailed analysis of autonomous cars with a particular focus on AI techniques used in the development of these cars. These cars use different AI techniques. Some of the well-known techniques are outlined below: Robots Basically, robotics is a field of artificial intelligence which deals with the study of robots. In this scenario, robots refer to machines that are used to carry out tasks which can be performed by humans. In fact, some robots are made so intelligent that they can perform their work without instructions. While the majority of robots always has an instructor guiding them what to do (Dunbar). Additionally, the robot is a Czech word, which is used for robota, meaning "forced labor." In addition, this term was first used by Czech writer Karel Capek in a play in 1920 for R.U.R.: Rossum's Universal Robots. In the play, the robots ultimately take over their creators (Rouse). Physically, a robot contains the same components as humans have. In this scenario, a robot has some kind of motor, a portable physical structure, a power supply, a sensor system and a processing system "brain" that is responsible for controlling all of these components. Fundamentally, robots are human created forms of animal life but actually they are machines which are intended to replicate human and animal behavior (Harris; Berlin). Artificial Neural Networks According to (Stergiou and Siganos), an artificial neural network is a data processing model that is inspired by the method of biological nervous systems information processing. Neural Network is also a renowned and heavily researched area of the AI. Stergiou & Siganos (2010) described that in the neural network field the working of human neuron is copied into the computer based systems. In addition, the neural network is an AI based computer program which imitates the human nervous systems’ working. It is based on neural net which has weights on each node that can be adjusted to make suitable decisions. Such systems are typically seen in automatic driving systems that are under-research and at the early stages of development (Stergiou and Siganos; Laudon and Laudon). Autonomous Cars An autonomous car, is also known using a number of other names such as self-driving cars, driverless cars or robot cars. Basically, these cars are designed using artificial intelligence techniques with the purpose of supporting the human transportation facilities which are provided by a traditional car. In view of the fact that it is an autonomous vehicle, hence it can sense its atmosphere and navigate without requiring human interaction. Though, until now these robotic cars are mostly used as demo systems and prototypes, however it is expected that in the future they will become more common. The working of these robotic cars is similar to artificial neural networks which are used using robotics. In this scenario, these cars use different components such as lidar, radar, computer vision and GPS to sense their environments. After that these sensory information is used by advanced control systems in order to translate and recognize suitable navigation paths, relevant signage and obstacles. In fact, some robotic cars are capable of updating their navigation paths and maps using these sensory information, permitting the cars to keep up to date with their location even when settings alter or when they enter the new atmosphere (Wikipedia; Aramburo and Trevino). Up till now, a large number of projects have been initiated to make intelligent vehicles. In this scenario, one of the most important project is initiated by Google. This project has attained a considerable success. In fact, now Google's self-driving cars have turned out to be open to the elements, rather accurately. Additionally, the company has started to test-drive them on public roads and even once they have invited public to drive one of the robot cars as it raced everywhere a closed course. Google's robot car that is known as Toyota Priuses has at the moment successfully completed the drive of approximately 300,000 kilometers (more than 190,000 miles). In fact, this car has been driven on busy highways, city traffic, as well as hilly roads with by requiring a least amount of human interference. Though, this success is still far from turning out to be practically possible, however the company has established a demo system on its location, by making use of self-directed golf carts. These golf charts can be used to take an idea of how these cars can be used to support human transportation facilities in the coming years. In addition, these details are still not published in any journal or book but these are discussed by Professor Sebastian Thrun, who is the project manager of this project, and engineer Chris Urmson in the form of a speech they delivered at the IEEE International Conference on Intelligent Robots and Systems in San Francisco last month (Guizzo). Figure 1Parts of the Google car, Image Source: (Guizzo) According to Google engineer Urmson the company has heavily relied on two most important approaches all the way through the development of this project. The first thing is that their technique depends on very thorough maps of the territory and roads. According to engineers, this is crucial to identify the exact position of the car. On the other hand, they have not relied on getting location using the GPS for the reason that locations attained through GPS-based techniques could be off by several meters. While the second approach is that, prior to sending the robot car for testing on a highway, their engineers drive along the road several times with the purpose of collecting information regarding the atmosphere. In this scenario, when it's the robot car’s turn to drive itself, it uses the previously collected data to compare the data that is valuable to distinguish pedestrians from static objects such as mailboxes and poles (Guizzo). There is another well-known example of robotic car known as Robot Car, which is developed by Department of Engineering Science at Oxford University in the UK. The project is undergoing under the batch of Mobile Robotics Group (MRG). However, this car can be distinguished from others on the basis of its cost in comparison to its technology. At the present, the project team encompasses more than 22 members in association with Nissan by using their hardware of the vehicle and technical support. In addition, Robotic Car is an amendment of the Nissan LEAF. The project team is intended to develop an environment-free navigation by means of autonomous perception and information engineering. Additionally, the project is intended to develop a safe driving car by linking with other drivers as well as making use of the infrastructures resourcefully at the same time as giving some free time to the drivers. Robot Car uses various AI techniques to keep an eye on the road, other cars and ramblers. In this scenario, if there is an obstacle onward, the speed of a car reduces itself and stops until the highway is clear again (Hulyacardas). Moreover, the car uses 3D sensing to start driving only after matching the stored 3D impression memory with the present atmosphere. In addition, like other AI technology this car first gets learning about its surrounding and after that it initiates the drive. This car encompasses three key components such as iPad, the Main Vehicle Computer (MVC) and the Low Level Controller (LLC). In this scenario, the iPad works as a user interface. Main Vehicle Computer is responsible for controlling everything about the drive. However, for a successful drive, all the three components must work in collaboration. Additionally, this autonomous car does not make use of the GPS, in its place it is planned like a robot which are used in logistics. Also, these are inexpensive as well as based on guide wires and reflective beacons (Hulyacardas). All the robotic cars work in a similar manner but they use different techniques or a mixture of techniques. Works Cited Aramburo, Jesus and Antonio Ramírez Trevino. Advances in Robotics, Automation and Control. Vienna, Austria: I-Tech, 2008. Berlin, D.S. Autoficial intelligence (2). 19 September 2011. http://www.economist.com/blogs/schumpeter/2011/09/driverless-cars. 20 November 2013. Dunbar, Brian. What Is Robotics? . 06 September 2013. http://www.nasa.gov/audience/foreducators/robotics/home/what_is_robotics_58.html. 18 November 2013. Guizzo, Erico. How Google's Self-Driving Car Works. 18 October 2011. http://spectrum.ieee.org/automaton/robotics/artificial-intelligence/how-google-self-driving-car-works. 20 November 2013. Harris, Tom. How Robots Work. 2013. http://www.howstuffworks.com/robot.htm. 19 November 2013. Hulyacardas. RobotCar the New Autonomous Car. 10 July 2013. http://www.roboticmagazine.com/military-medical-vehicles/robotcar-the-new-autonomous-car. 23 November 2013. Laudon, Kenneth C. and Jane Laudon. Management Information Systems: Organisation and Technology in the Networked Enterprise, 6th Edition. New York: Pearson, 1999. Rouse, Margaret. robot (insect robot, autonomous robot). April 2007. http://searchcio-midmarket.techtarget.com/definition/robot. 20 November 2013. Stergiou, Christos and Dimitrios Siganos. "NEURAL NETWORKS." 2010. 14 November 2013. . Wikipedia. Autonomous car. 30 November 2013. http://en.wikipedia.org/wiki/Autonomous_car. 30 November 2013. Read More