Self-Driving Cars - Research Paper Example

Background Information Self-driving or robotic cars are automated driverless cars that have the capability of fulfilling the main transportation expectations just like traditional cars. These cars can sense their environments using techniques such as lidar, radar, computer vision and Global Positioning System (GPS) and respond appropriately without the human input (Ferris, 2011). Autonomous cars have existed mainly as prototypes and system demonstrations over a long period. They interpret the information using their advanced sensory system to identify appropriate navigation paths, obstacles and relevant signage. The autonomous cars are also capable of updating maps based on sensory input thus allowing them to keep track of their location even when the conditions change or when they navigate uncharted environments. Houdina radio control n 1925 a radio controlled driverless car on New York’s Fifth Street Avenue amidst a traffic jam. The car, Linrrican wonder was an improvement of 1926’s Chandler fitted with transmission antennae. Another car from behind acted as a control base that transmitted signals to the receiving antennae. Circuit breakers carried the signals and operated small electric motors responsible for controlling the car. In 1939, a General Motors (GM)’s sponsored exhibit, Futurama designed by Norman Bel Geddes, was unveiled at the world Trade Fair. The electromagnetic fields generated by the embedded circuits in the highway provided propulsion while control was radio based. Bel later outlined his mission in a book entitled Magic Motorways in 1940. These developments prompted the improvements in highway design and transportation. In 1923, Radio Corporation America (RCA) labs successfully built a miniature controlled by wires laid on the laboratory floor. It was an experimental system that jerked the imagination of a traffic engineer in Nebraska called Leland Hancock (Jurgen, 2013). Leland and his director, therefore, decided to experiment this system with actual highway installations. In 1958, RCA Labs and State of Nebraska successfully demonstrated Leland’s experiment on a four hundred foot public highway outside Lincoln’s Neb (Le Vine, Zolfaghari & Polak, 2015). Many experimental detector circuits buried in the pavements electric impulses to guide the car and determine the presence of speed (velocity) and of any metallic vehicle moving on the surface in collaboration with the General Motors. In 1960s, Communications and Control Systems Department of Ohio State University launched a driverless car project controlled by electronic devices embedded in the highways. These developments prompted the Bureau of Public Roads to consider constructing an experimental electronically controlled highway (Lee, Lee, Yoon & Cho, 2013). Ohio, New York, California and Massachusetts, therefore, forwarded their bids for the highway construction. In 1961, William Bertelsen invented an aero-mobile 35B, an air-cushioned vehicle envisioned to revolutionize transport system with personal auto-driving hovering cars with a speed limit of 1500 mph. During the same time, London under United Kingdom’s Transport and Roads Research Lab tested Citron Ds, a driverless car that interacted with magnetic cables embedded in the roads. Citron cruised through a test track at a steady speed of 130 kilometers per hour without signs of deviation in any weather condition. In 1980s, Ernst Dickmann and his team of engineers in Bundeswehr University Munich in Germany designed a vision-guided Mercedes Benz robotic van (Liniger, Domahidi & Morari, 2014). The van moved at a speed of 63 kilometers per hour on the streets without traffic jam. EUREKA also conducted a multi-million Prometheus Project on autonomous vehicles from 1987 to 1995. In the same decade, Autonomous Land Vehicle (ALV) in the United States employed the new control technologies was developed. A joint project conducted by Carnegie Mellon University, the University of Maryland, Environmental Research Institute of Michigan, Martin Marietta and SRI International achieved the first road demonstrations. The car used computer vision, lidar, and robotic autonomous control for direction at a speed of 31 kilometers per hour (Marks, 2012). Hughes Research Laboratories (HRL) Laboratories set the record by demonstrating the first off-road map and sensor-based autonomous navigation on ALV in 1987. The vehicle travelled a distance of 610 meters at a speed of 3.1 km per hour in a complex terrain of ravines, steep slopes, large rocks and vegetation. The United States Congress passed the Transportation Automation Bill in 1991 that instructed the United States Department of Transportation (USDOT) to demonstrate self-driving vehicles and highway systems by 1997 (Münch, 2014). The Federal Highway Administration took the challenge by first conducting series of precursor system analyses followed by the establishment of National Automated Highway System Consortium (NAHSC). The project was a cost shared by the General Motors, Caltrans, and Delco companies among others. The extensive research on engineering resulted in a major demonstration in San Diego, California involving twenty automated vehicles including cars, trucks and buses. This public demonstration comprised a close headway-platooning intended to operate on separate traffics, as well as free agents vehicles operating in a mixed traffic. Many carmakers companies such as Toyota, Honda among others participated in the event with an aim of producing a system design that will help in the commercialization of driverless cars (Pettersson & Karlsson, 2015). However, the program ended in late nineties due to tight research budgets at USDOT. 1n 1994, VaMP and Vita-2, twin robot of Daimler-Benz and Ernst Dickmann cruised over 100 kilometres in a Paris’ three-lane highway amidst a standard heavy traffic. They cruise at an average speed of 130 kilometres per hour semi-autonomously with minimal human interventions. These machines also demonstrated self-driving in free lanes, convoy driving as well as lane changing and autonomous overtaking of other cars. In 1995, Navlab project of Carnegie Mellon University completed a 5000 kilometers cross-country in an operation dubbed No Hands across America of which 98.2% of the journey was self-controlled. The semi-autonomously controlled car had only throttle and brakes under human control while the vehicle via neural networks autonomously controlled the steering wheel. The re-engineering of autonomous Mercedes-Benz S-Class in 1995 by Dickmann completed a 1590 kilometers journey in Munich, Germany to Copenhagen in Denmark and back. The car used a saccadic computer vision and transputers for real-time reactions. The car achieved an average speed of about 175 kilometers per hour on the German Autobahn with a 0.9-kilometre mean time human intervention (Singh, 2014). The machine maneuvered the traffic successfully including overtaking through self-execution and drove up to 158 kilometers without human intervention despite being a research system. In 1996, a University of Parma professor, Alberto Brogi launched an AGRO project, a modification of a previous project, Lancier Thema, to follow lane marks in an unmodified highway. The vehicle equipped with only two black and white video cameras used stereoscopic vision algorithms to understand its environment ('Two billion cars: driving toward sustainability', 2010). The project car covered a 199 kilometers journey with an average speed of 90 kilometers per hour. It fully operated in an automatic mode for 94% of the entire journey. Production vehicle when they launched a laser-based system for the compact luxury sedan, later sold to Japan. In 2001, the American government funded three military efforts known as Demo I, ii and iii that demonstrated the ability of unmanned ground vehicles to navigate many kilometers of difficult off-road terrain. James Albus provided a real-time control system in an experiment at the National Institute of Standards and Technology. The system not only controlled individual vehicles, for example, throttle, steering, and brakes but groups of vehicles had their movements coordinated automatically in response to high-level goals. In 2007, the Defence Advanced Research Projects Agency (DARPA) sponsored a challenge for driverless cars in an urban environment where Chevy Tahoe, an autonomous car from Carnegie Mellon University claimed the first place (Urmson & Whittaker, 2008). Such events gave students and researchers opportunity to research on self-driving cars that work towards reducing transportation burdens and problems such as traffic jams and accidents that are on the increase in many urban centers. By early 2000s Park Shuttle, driverless public road transport system started operating in Netherlands. In December 2008, Rio Tinto Alcan tested Komatsu autonomous haulage in Pilbara iron ore mining site in Western Australia. The system became the first world’s commercial autonomous mining haulage system. This system has come with benefits in terms of health, safety, and productivity. From the year 2010, a major technological showdown in autonomous cars among world’s major automotive manufacturers heightened. These companies include Ford, Mercedes Benz, GM, Volkswagen, Audi, BMW, Toyota, Nisan, and Volvo, who independently developed the models. BMW has been testing a self-driving car system since 2005 while Audi sent a driverless car, Audi TTS atop of Pike’s Peak in 2010 at a close to race speeds. GM created an autonomous electric urban vehicle called Electric Networked Vehicle (EN-V) in 2011. Volkswagen started testing a Temporary Auto Pilot (TAP) system in 2012 that will allow a car to auto drive at a speed of 130 kilometers per hour on a highway (Yang & Coughlin, 2014). Ford has done a comprehensive research in driverless systems, as well as vehicular communication systems while Toyota demonstrated a partially driverless car in January 2013 with numerous communication systems and sensors. The Freie University in Berlin developed two self-driving cars tested in the inner city traffic of Berlin in 2011. The two systems successfully handled the inner city traffic, traffic lights as well as roundabouts. A 22km driving test administered to a Google autonomous car by Nevada motor vehicle examiners in a test route in Las Vegas City proved successful ('Autonomous Vehicles: A railway perspective', 2013). The car passed the test without examination on roundabouts, signals, rail crossings, and school zones. July 2013 marked another pioneering test by VisLab driverless vehicle that successfully drove in Parma downtown with no human control through roundabouts, pedestrian crossings, traffic lights and other common hazard signs. Daimler R&D in collaboration with Karlsruhe Institute of Technology made a Mercedes Benz S class driverless vehicle in 2013. The system fitted with radars and stereo cameras completely drove autonomously for about 100 kilometres from Mannheim to Pforzheim in Germany. Nisan announced plans to launch many driverless cars by 2020. The company has just completed building an autonomous driving proving ground in Japan. For demonstration purposes, the company installed autonomous car technology in a Nissan Leaf electric car that participated in a 360 test-drive event held in California in 2013 (Iozzio, 2015). Nissan engineers use this testing car to evaluate how this autonomous driving software performs in the real world. A 2014 Mercedes S class has alternatives for self-governing steering, acceleration and braking, lane keeping, accident avoidance, driver fatigue detection, and parking. These are applicable both in the city traffics and at highway speeds of up to 200 kilometers per hour. The 2014 Infiniti Q50 fitted with cameras, radar, and other sophisticated technology designed for lane keeping, collision avoidance and speed control features. Many contemporary self-driving cars have features offering limited autonomous functionality (Iozzio, 2015). These features include adaptive speed control, a system monitoring distances between adjacent vehicles in the same lane, auto-adjusting the speed depending on the traffic flow. Lane assists technology monitors the position of the vehicle within the lane and alerts other vehicles when it is leaving its lane while parking helps in parallel parking. Navia shuttle of induct technology became the first autonomous vehicle available for commercial sale in January 2014. With a 20 kilometres per hour speed limit, the open-air electric vehicle that resembles a golf cart can accommodate eight people. It is intended to transport people around in city centers, large industrial sites, airports, hospital complexes, theme parks or university campuses. Google also plans to unveil almost 100 self-driving car prototypes built from their secret X labs. In October 2014, Telsa Motors announced the first version of AutoPilot cars. These Model S cars fitted with systems that are capable of autonomous steering and lane control with braking and speed limit adjustment based on signals or image recognition. The system is also capable of packing autonomously, as well as receiving software updates to improve the skills. Testing of this system commenced on March 2015 on the highway between Seattle and San Francisco with the car driving almost unassisted (Achtenová & Korec, 2011). At the same time, Telsa Motors announced their intention to introduce their Autopilot technology through a software update for all the cars equipped with the autonomous driving system. Advantages In 2015, the United Kingdom government launched public trials for the driverless pod called LUTZ Pathfinder in Milton Keynes. Commuters considering themselves to be early adopters will be tooling around in autonomous cars before 2030 with mass adoption likely to occur by 2040 (Edwards, Nathanson & Wisch, 2014). The benefits of self-driving cars are not just for themselves, but the entire society. Since these cars can easily drop off passengers and park themselves, parking spaces would not need to be wide enough to allow for the opening of doors. The move could free up billions of square yards from the current parking lots. The freed up area can generate additional revenue when used for other developmental purposes thereby contributing to the increase in the GDP. The additional revenue will translate to improvement in economic growth and the people’s living standards. Self-driving cars can eliminate human driver error and are unlikely to cause accidents. Various World Health Organization reports estimate global deaths related to motor vehicle accidents to be 1.24 million annually with around 2 Million related injuries. The major cause of these accidents is human error, a problem solved by self-driving cars (Brombacher, 2014). These cars would lower the deaths and injury rates that result from such accidents thereby saving the people from unnecessary expenses. People’s savings will increase, as well as the economic growth in general. The US suffers a cost of up to $625 billion annually due to motor relate accidents. Since 90 percent of motor vehicle crashes results from a human error, substituting drivers with driver-less cars could reduce the cost of accidents by 90 percent thus saving the country about $563 billion per year (Achtenová & Korec, 2011). In comparison with the many bad behaviors exhibited by most of the drivers behind the wheel, a computer, therefore, proves to be an ideal motorist. Since 90 percent of the car crashes results from a human error, computers are unlikely to cause the same. There is also no question or opportunities of computers getting distracted by events, scenes or thoughts (Jurgen, 2013). Although it is unquantifiable to determine the extent of life saving, it is obvious that the human controlled cars carry very high risks in terms of danger. Computers use complicated and precise technology based on algorithms to determine appropriate stopping distances, the distance from other vehicles and other information or data that drastically lowers the cars’ chances of accidents (Achtenová & Korec, 2011). The result is a significant saving in various avenues such as insurance and healthcare. Insurance companies would no longer be insuring motor vehicle drivers against accidents resulting from their mistakes but would instead switch their models to insure the self-driving cars against technical failure. Insurance premiums would, therefore, plummet thus saving people money. Many people will opt to share cars or use autonomous taxi thereby greatly reducing road congestion. The result will greatly reduce traffic snarls thus save people from time spent in the traffic jams resulting in cost saving in terms of time. When a computer assumes the driving role, drivers can use that time to do other things that contribute to the economic productivity of the nation such as online business, networking or even chat with the passengers (Ferris, 2011). People will also be able to work in their cars en-route to work, hold meetings while travelling, or do any recreational activities thus improving their productivity. Fatigue from driving in long distance journeys will be outdated with the revolution of road transport. A large number of Self-driving cars contributes to a behavior called platooning. This behavior significantly improves the conditions of the traffic in terms of congestion. The effect will help in reducing commuter hours for drivers in heavy traffic areas. For the cars to operate, they need to communicate with each other thereby helping in the identification of the possible traffic problems or road risks early enough (Brombacher, 2014). They are also able to understand their environments thus enabling them to identify the optimum route. Driver-less vehicles are also capable of communicating with roadside infrastructures such as traffic lights and signs and use this information to minimize fuel consumption, as well as emissions. People living with disabilities who have to rely on public transportation or assistance by other people to get around could enjoy the benefits of self-driving cars (Jurgen, 2013). This new enhancements and freedom of mobility are none discriminative hence suitable to all members of the society regardless of their physical status. Driverless cars will also offer an opportunity for the blind to drive. The physically disabled would, therefore, drive themselves easily on these cars without employing drivers. Big cities face a crisis of providing adequate transportation system. Many of these cities lack the appropriate facilities and infrastructure that can sufficiently support their residents’ needs. The self-driving vehicles will, therefore, fill this void since they economically use parking lots and are easily accessible to everyone regardless of age, disability status or whether one knows how to drive or not. They also make the cities neat and clean due to organized computerized parking. Many companies are interested in new product developments that hit the market and take the industry forward, a satisfaction offered by the driverless cars. Higher speed limits will be an option if self-driving car usage becomes popular since the computers accurately calculate the vehicle's operation safety. Driving hours will reduce by faster speeds allowed on the roads. The development increases the possibility of covering great distances within a short time thus increasing mobility of people and goods. Trade will, therefore, improve (Ferris, 2011). These cars also give drivers hints on obstacles in the environment because they have sensors that give environmental sensitivity as well as gadgets that initiate autonomous braking, speed control, and self-parking. Drunk driving incidents would decline dramatically since no designated driver is needed when the car auto-drives itself. Massive savings from older mass transit projects like the train would accrue. The police offer focus could also shift from writing traffic tickets, handling accidents and operating roadblocks to managing crimes that are more serious (Althoff & Mergel, 2011). The sensors in the autonomous cars allow them to ride closer to one another thus allowing more car population on the road with less traffic. There would be less parking structures and parking headaches since the car can drop one off and locate a parking space elsewhere. People would not need a specialized driving license to operate the cars. It will also be needles to take the key away from grandpa when he gets old to drive carefully, the car will take care of that. Today’s cars in America, when using cruise control and smooth driving, can deliver fuel economy saving of up to 30 percent. Self-driving cars will, therefore, be more fuel-efficient since they will always be in cruise control 100 percent of the driving time. This factor together with their improved aerodynamics, lighter weight materials and other technological advancements enabling the authors predict precisely (Brombacher, 2014). These cars can be 30 percent more efficient and economical with fuel consumption than an equivalent non-autonomous car. The is a reduction in the nation’s expenditure on gasoline bill. Thus, this savings can boost other sectors of economic development. Disadvantages Despite the many benefits that come with the self-driving cars, the shortcomings of this technology are equally abundant. In the same way knowledge is required to drive the normal cars, these autonomous machines require knowledge to operate them too. Regardless of the computer taking control once the vehicle is in operation, it would still need a driver with knowledge about it since it cannot read the human mind (Irle, Gröll, & Werling, 2009). The driver has to feed instructions that the vehicle will automatically execute, for example when to stop. Implementing the new technology would be costly hence many people cannot afford thus the benefits of autonomous cars will just remain a dream to them. If adoption of this technology falls short of the expectations, the savings in terms of cost, lives, and time will remain a tall order (Brombacher, 2014). Accidents, deaths and other related injuries can still happen just like before. A total elimination of accident with the adoption of self-driving cars is uncertain. There is also no legal jurisprudence set to handle cases of such accidents. It is also difficult to hold anyone responsible in case of an accident since the manufacturer, owner or driver, and software developer would counter accuse one another. For a computer software to operate a vehicle, a huge amount of information would have to be stored in the software. Some people who are more concerned with the opportunity for a computer built in the autonomous cars are likely to hack into the system, re-program it and steal personal data. They can also reprogram the software to cause harm or accident to the occupants. If the computer crashes or breaks down, repairing the car will require highly qualified personnel who understands the system and such people are hard to come across. Failure of other technologies such as traffic light signals that the cars depend on would mean no accounting for human traffic signals. In an event of an accident where the traffic police redirect the traffic, the cars cannot interpret human signals thus are likely to course more accidents. Self-driving cars would render many people jobless by eliminate many jobs in the transport sector, especially when it comes to taxi and freight transportation. Many mechanics would be jobless due to the complexity of the new technology, passenger service drivers, traffic police officers as well as the parking attendants. This decline in income would greatly affect the economy negatively. Over-reliance on self-driving technology could mean that with time, drivers will be devoid of the basic driving knowledge to operate normal cars. In the event of a technology failure, drivers will be helpless to get around in the manual way. The cars are also limited to perform at the high level of safety in extreme weather conditions (Althoff, 2010). Heavy rains can do serious damage to the laser sensors mounted on the roof of the car raising questions on the driver’s role in such events. The self-driving cars rely on accurate systems mapping via GPS. GPS devices are not always accurate, for example, GPS may advise anyone to turn down a two-way street or that they are driving on a non-existence street. These confusions are raising security concerns about these vehicles. It is also unclear how these cars will maneuver roadblocks and unique driving laws. For example, the difference that exists between states regarding turning right on red. The computers could have difficulties in identifying different locals and state rules regarding the road. Issues The issues that have surrounded the driverless cars over time are key to the adoption of this technology by the people. Some of them include the question of autonomous car liability. These have been tricky, but a near-consensus possibility is that insurance companies will embrace them due to the total elimination of human driver error. Other contentious issues include whether sitting behind the wheel will require a driver’s license or not and amending the traffic and transport rules to accommodate the driverless cars. Conclusion Driverless car technology is a potential game-changer on the world roads, altering the face of motoring in a very fundamental way and delivering greater benefits to road safety, emissions, social inclusion, congestion and economy. Despite the challenges, the technology remains the most superb for both speed and convenience. References Achtenová, G., & Korec, O. (2011). Autonomous driving algorithms in scaled environment. Journal of Middle European Construction And Design of Cars, 9(3). doi:10.2478/v10138-011-0016-y Althoff, M. (2010). Reachability analysis and its application to the safety assessment of autonomous cars. Althoff, M., & Mergel, A. (2011). Comparison of Markov Chain Abstraction and Monte Carlo Simulation for the Safety Assessment of Autonomous Cars. IEEE Trans. Intell. Transport. Syst., 12(4), 1237-1247. doi:10.1109/tits.2011.2157342 Autonomous Vehicles: A railway perspective. (2013). Engineering & Technology Reference. doi:10.1049/etr.2013.9008 Brombacher, A. (2014). (Re) liability of Self-driving Cars. An Interesting Challenge!. Qual. Reliab. Engng. Int., 30(5), 613-614. doi:10.1002/qre.1707 Edwards, M., Nathanson, A., & Wisch, M. (2014). Estimate of Potential Benefit for Europe of Fitting Autonomous Emergency Braking (AEB) Systems for Pedestrian Protection to Passenger Cars.Traffic Injury Prevention, 15(sup1), S173-S182. doi:10.1080/15389588.2014.931579 Ferris, T. (2011). The big idea. Washington, D.C.: National Geographic. Iozzio, C. (2015). Driverless Tech Inches Ahead. Sci Am, 312(4), 16-16. doi:10.1038/scientificamerican0415-16 Irle, P., Gröll, L., & Werling, M. (2009). Zwei Zugänge zur Projektion auf 2d-Kurven für die Bahnregelung autonomer FahrzeugeTwo Approaches for the Projection onto 2d-Curves for the Path Control of Autonomous Cars. At - Automatisierungstechnik, 57(8). doi:10.1524/auto.2009.0785 Jurgen, R. (2013). Autonomous vehicles for safer driving. Warrendale, Pa. (400 Commonwealth Dr., Wallendale PA USA): Society of Automotive Engineers. Le Vine, S., Zolfaghari, A., & Polak, J. (2015). Autonomous cars: The tension between occupant experience and intersection capacity. Transportation Research Part C: Emerging Technologies, 52, 1-14. doi:10.1016/j.trc.2015.01.002 Lee, S., Lee, J., Yoon, K., & Cho, H. (2013). Intelligent Autonomous Systems 12. Berlin, Heidelberg: Springer Berlin Heidelberg. Liniger, A., Domahidi, A., & Morari, M. (2014). Optimization-based autonomous racing of 1:43 scale RC cars. Optimal Control Applications and Methods, n/a-n/a. doi:10.1002/oca.2123 Marks, P. (2012). Autonomous cars ready to hit our roads. New Scientist, 213(2858), 19-20. doi:10.1016/s0262-4079(12)60813-6 Münch, B. (2014). Legal Questions with autonomous Cars. Saarbrücken: AV Akademikerverlag. Pettersson, I., & Karlsson, I. (2015). Setting the stage for autonomous cars: a pilot study of future autonomous driving experiences. IET Intelligent Transport Systems. doi:10.1049/iet-its.2014.0168 Singh, R. (2014). Driverless Cars: Truly ‘Auto’Mobile. Auto Tech Rev, 3(10), 12-13. doi:10.1365/s40112-014-0756-x Two billion cars: driving toward sustainability. (2010). Choice Reviews Online, 47(06), 47-3292-47-3292. doi:10.5860/choice.47-3292 Urmson, C., & Whittaker, W. (2008). Self-Driving Cars and the Urban Challenge. IEEE Intelligent Systems, 23(2), 66-68. doi:10.1109/mis.2008.34 Yang, J., & Coughlin, J. (2014). In-vehicle technology for self-driving cars: Advantages and challenges for aging drivers. International Journal of Automotive Technology, 15(2), 333-340. doi:10.1007/s12239-014-0034-6 Read More