Until the 2000s, self-driving cars merely belonged to novels, movies and tales. Various sci-fic writers – like Isaac Asimov – held reader’s attention with breath-taking elements such as autonomous cars that communicated, fought and showed their feelings. They have been doing so since the 1930s.
However, these dreams and products of imagination are fast becoming a reality with various self-navigating being tested. The hopes of having self-driving cars on roads were further rekindled between 2004 and 2012 when huge progress was recorded in the automotive industry.
Self-Driving Car: An Overview
What is Self-Driving Car?
A self-driving car is a vehicle that has the ability to sense its surrounding environment and navigate without any contributory human efforts. It is a vehicle designed to self-navigate or travel without an operator. What qualifies as a self-driving car is a vehicle that is able to navigate to the desired destination, without human intervention and on roads that have not been adjusted for robotic cars.
Also known as an autonomous or robotic car, the self-driving car uses various methods to recognise its surroundings. These methods include laser light, GPS, radar, computer vision and odometry.
From Google explorations, a self-driving car operates in an easy way. When travelling by the autonomous car, a user sets a destination then the software in the car’s operating system determines the route, kickstarts the car and begins driving towards the intended destination.
The robotic car is designed to feature a radar mechanism on its rear and front bumpers which estimates the distances to obstacles. The vehicle features a rotating LIDAR (Light Detecting Ranging) sensor monitor that provides a 3-D map of the car’s environment. The vehicle features a sensor on the rear wheel that monitors movements for detecting the position of the car in relation to the 3-D map.
Google designed its robotic vehicle in way that it is compatible with Google Street View and Google Maps. The Artificial Intelligence in the car connects with these Google interactive systems for notification about landmarks, traffic lights and signals. The interior of the vehicle features video cameras and an override function to allow a user take full control whenever necessary.
There are people who strongly believe that self-driving cars can eliminate road accidents. Robotic cars are developed to avoid human errors that led to car accidents. Also, robotic cars are the veritable solutions to traffic jams and they are likely improvements for highway capacity. However, an automatic vehicle allows for easy commuting as travellers can relax, read or sleep.
Classification of Self-Driving Car
In 2014, SAE (Society of Automotive Engineers) International published a classification system containing six separate levels of self-driving car. The automotive standardisation body introduced this classification system based on the level of the driver intervention and attention need other than the abilities of the vehicle. These classifications are mentioned below:
This level contains systems vehicles with automated systems to give warnings but do not have total control.
Vehicles in this level share navigation control with drivers. The level is also called ‘hands on’ level; one of the example of vehicle in this level is such where the driver exercises control over steering while automated system handles speed. This system is more commonly known as Adaptive Cruise Control (ACC).
Parking Assistance, where the automated system controls the steering while the driver controls the speed, is another example of level 1 vehicle. Also, Lane Keeping Assistance (LKA) Type II is another example of level 1 driving system.
Generally, this system demands that the driver is at qui vive to control of the vehicle at this level.
This grade of automated system includes vehicles with the ability to take control of acceleration, brake and steering. Cars in this level takes full control of the vehicle. this SAE driving level is also called ‘hands off’. However, the driver must pay close attention to the vehicle and be ready to take control of the vehicle when it is not responding appropriately.
This level is equally called ‘eyes off’. Automated vehicles in the range are equipped to allow drivers to pay attention to other things. For example, they can send a text message or read an article. A level 1 self-driving car can conveniently handle emergencies by applying emergency brake. However, vehicles in this category still requires some form of human intervention. Users of level 3 vehicle must read the vehicle manual to know how to handle the vehicle whenever an intervention is needed.
The vehicle allows the driver to do other things. The driver can safely sleep or completely leave the driver’s seat. The vehicle in this grade is designed to handle traffic jams, park and stop.
This level is the highest in the SAE category. Vehicles in this grade do not require human intervention. Already, Uber has introduced robotic taxis of this kind.
Numerous Concerns About Self-Driving Car
Many people are yet to wrap their heads around what self-driving car really is. Truly, the future is uncertain about what a self-driving car will look like. For some, it is even doubtful that the self-driving car technology will see the light of the day.
There are legal concerns; as several individuals nurse the fear of losing complete control of automation to robots. Also critics are concerned about the safety of lives and protection of privacy as self-driving operating systems could be hacked.
Component Parts of Self-Driving Car
Self-driving car is built on important technological components. Modern robotic vehicles are powered by Bayesian SLAM (Simultaneous localization and mapping) algorithms for fusing data taken from multiple sensors with an off-line map into the present location estimates and for updating map.
There are indications that modern self-driving cars will also feature DATMO (detection and tracking of other moving objects) for recognising vehicles and pedestrians. This feature is especially linked to models from Google.
Also, self-driving cars feature RTLS (real time locating system) systems to help the system of the vehicle to understand the surrounding environment. It features GPS and stereo vision as well.
Historical Development of Self-Driving Car
The Conceptualisation of Self-Driving Cars
The self-driving car was first conceived in the medieval period. There are traces to one of Leonardo da Vinci’s works, where he sketched an autonomous cart. But of course, the idea he conceived was not as advanced as those being nurtured in the minds of technologists and inventors today.
The first real attempt to invent the self-driving car began in the early 20th century. In 1925, the Houdini Radio Control Company demonstrated a driverless car. It was a 1926 radio-controlled Chandler. The vehicle was directed through a road along the Fifth Avenue and Broadway from signals from near-by vehicle. Although the vehicle was not directly driven by a human, it needs an external influence for propellant.
Not long after, Achen Motor demonstrated the Phantom Auto in Milwaukee. It was a remote-controlled vehicle that stunned onlookers at the demonstration grounds. The year was 1926.
The Phantom Auto ended up on numerous tour from the ’20s to ’30s. It drew mammoth crowd, as people wanted to see if there was an iota of truth in the claims that a vehicle could be remote-controlled. However, Phantom Auto, like Chandler, was flawed. Although driverless, it was not autonomous as it needed to be controlled from afar.
Self-Driving Cars for Smarter Future
In 1939, the world witnessed a more robust demonstration of what the future could hold as regards transportation. Norman Bel Geddes’ exhibition of ‘Futurama’ at the World’s Fair astounded the participants at the fair. The industrialist presented innovative and realistic ideas of transportation in 20 years from the time.
One of the major components of Bel Geddes’ transportation was expressways as a way to connect cities and communities. It also projected highways with autonomous cars. Geddes’s idea about autonomous cars was such that they could become a safe means of transportation for commuters. They could arrive at their destinations safely and quickly.
According to Bel Geddes in Magic Motorways (his 1940 book), “These cars of 1960 and the highways on which they drive will have in the devices which will correct the faults of human beings as drivers.”
Quickly, the Radio Corporation of America (RCA) collaborated with the state of Nebraska and General Motors to develop Bel Geddes’s idea. They started developing automated highway technology in line with Bel Geddes’s concept. The crew introduced an automated highway with electronic pavements. The structure was 400-foot long, and the electronic circuits in the pavements were used for measuring varying road states and propel vehicles along the road parts. The test was a success and the second version of the scientific work was demonstrated in Princeton, New Jersey, in 1960.
The success of the works encouraged the RCA so much that they revealed their plans to commercialise the technological works on a large scale in the next 15 years. Consequently, General Motors manufactured and launched a fleet of electronic cars that were fitted for the pipeline smart roads. The cars, Firebird II and Firebird III were heavily advertised. The vehicles were characterised by their idealist and futuristic outlook. They had a high-tech guidance mechanism wired to work in unison with the electronic circuits on the highway.
Unfortunately, the US government did not key to the idea of futuristic cars. This lead to the halt in progress on the development of a smart highway. The US government was required by the RCA to invest a whopping $100,000 a mile fund to make the larger version of the highway.
However, the United Kingdom’s Transport and Road Research Laboratory (RLL) were working towards their version of self-driving cars. The RRL were working on a technology that was close to the automated highway mechanism of the US. Their system also involved a car and road mechanism. In this instance, combined a Citroen DS, upgraded with electronic sensors, with a magnetic rail track that was buried under the road.
In spite of the series of successful tests and impressive projections that the project would drastically reduce road accident, increase the capacity of city roads and become highly profitable, the UK government stopped funding the project.
More Explorations on Self-Driving Cars
Yet, organisations and researchers have not given up the dream of self-driving cars. The project is truly costly but the dream is bigger to be dropped. Several attempts were made in the ’60s to introduce more affordable self-driving cars.
Resultantly, researchers started focusing on automated cars only, in lieu of the earlier developments that included the highways. They needed to make the cars alone starter and work on the traditional roads. The first set of engineers to key into this scope were those at Stanford.
These graduate students were keen on changing the world. James Adams was the first to blaze the trail. He coupled a four-wheeled cart in his first attempt to develop a loner rover propelled by remote control. The cart featured a video camera, for improved navigation. The design coalesced into a more developed and intelligent vehicle that could self-navigate. There were several experiments with which the vehicle successfully drove past sets of chairs in a room.
In addition, a crew at Tsukuba Mechanical Engineering Laboratory, Japan, ventured into the development of self-driving vehicles. Their 1977 attempt has been regarded as the first major step to inventing a stand-alone development of self-driving cars. The invention did not include road technology. Their creation relied on the machine vision, a series of built-in cameras, to analyse its surroundings for guidance.
The model had the ability to cover 20 miles per hour. The self-driving vehicle was programmed to move in tandem with white street markers.
It could be said that the invention deepened interests in artificial intelligence. Artificial intelligence was heavily applied to transportation, growing incrementally in the ’80s. Ernst Dickmanns, a German engineer, spearheaded the application of AI to transportation in the ’80s.
He collaborated with Mercedes-Benz to upgrade a Mercedes van using sensors and cameras. The vehicle was able to self-regulate its steering, brake, wheel and throttle through a computer application; the application is programmed to receive and transfer data to the van. The Mercedes van is tagged VAMORS; was successfully demonstrated, and was launched.
Bigger Explorations and Projects
The successful launch of VAMORS encouraged the European Research Organisation (EUREKA) to develop and introduce the Prometheus project. This project is considered the most ambitious attempt into the development of self-driving cars. It involved an investment of over 700 million euros and various researchers.
Ernst Dickmanns was involved in the project. He, alongside Bundeswehr Universitat Munchen researchers, developed major camera technology, computer processing and applications that facilitated the invention of impeccable smart cars. VaMP and VITA-2 were tested and launched. The two vehicles were made to travel on a 1,000-kilometre highway at the speed of 130 kilometres per hour.
Researchers in the US was also embarking on a series of experiments on self-driving car technologies. For instance, researchers at the Carnegie Mellon Robotics institute tested some fleet of cars. They began with NavLab 1, a Chevrolet van. The car was upgraded with some video gadgets, a computer and a GPS receiver. Researchers at Hughes Research Labs also demonstrated a version of a self-driving car that can self-navigate.
Soon after, a team of researchers at the University of Parma, led by professor Alberto Broggi, commenced with the ARGO project. The project is aimed at developing on EUREKA’s abandoned Prometheus project. What this meant is that cars could become totally autonomous and affordable. They wanted to fully functional and compact self-driving cars, so they developed Lancia Thema. The car had a black-and-white camera and stereoscopic vision algorithms based navigational system. Lancia Thema was noted to be very speedy, covering a distance of over 1,200 miles at a remarkable speed of 56 miles per hour.
Furthermore, the US military announced its DARPA Grand Challenge, beginning in the 21st century. The competition promises to award a team of engineers whose car is capable of surmounting the 150-mile record speed of the autonomous vehicle. Although there was no winning vehicle, the competition encouraged more explorations in the autonomous vehicle technology industry. It propelled innovations in the industry as similar competitions were held in later years.
Self-Driving Cars in Recent Times
In 2010, Google, the largest search engine organisation, announced that it had explored the path to self-driving cars the year before. The leading internet company had some of its employees develop and test a mechanism for a self-driving the car. The goal was to create a solution to the problems of road accidents and reduce road carnage by half ever year.
Sebastian Thrun; director of Artificial Intelligence Laboratory, Stanford, led the project. He and his team of engineers (most of whom worked on cars that were used to compete in the DARPA Grand Challenge competitions) were charged with the responsibility of launching a commercial self-driving vehicle in 2020.
So far, the team has been able to get legal backing from the US government on the legality of self-driving cars on street. Self-driving cars were declared legal in four different states.
The team first tested seven models, six Toyota Prius cars and one Audi TT. The vehicles were upgraded with a series of lasers, advanced radar, cameras and GPS technologies. These gadgets enabled the vehicles to perform better than merely circumnavigating a route. These smart vehicles can recognise hindrances from afar. They are sensitive enough to recognise people and roadblocks.
In 2015, Google self-driving cars had recorded over 1 million miles without major accidents. There were only 13 collisions that the vehicles were involved in. Last year, there was the one incident that a major accident was recorded, this was caused by a mistake caused by the smart vehicle.
The works of Google inspired the District of Columbia to make provision for smart vehicle transportation. The district introduced a 100 per cent autonomous prototype, to be released in 2020. Similarly, numerous automobile and transportation companies are investing in self-driving cars technology to further the advancement of self-driving cars in the nearest future. Some of these companies are Tesla, Toyota, BMW and Uber.
Possible Advantages of Self-Driving Car
Considering the prospects of a self-driving car; its likely component parts and technology; there are possible advantages that can be ascribed to a self-driving car.
Reduction in Road Accidents
One major possible advantage of a self-driving car is that it can reduce road accidents. Road carnage caused by errors of human drivers causing deaths, injuries, destruction of social amenities and waste of funds can be drastically reduced or eliminated.
While human drivers can make mistakes like delayed reaction, rubbernecking, tailgating and give in to numerous distractions on the road or drive aggressively, robots are not expected to make similar mistakes.
McKinsey and Company, a consulting firm, calculated that self-driving cars is capable of eradicating 90% of all auto accidents in the United States. According to the firm, robotic cars can avoid close to US$190 billion in damages, loss of lives and huge costs invested in annual health costs.
Reduction in Labour Costs, Waste of Man Hours and Fuel
Similarly, self-driving cars can help avoid labour costs. The technology is capable of relieving commuters of stress related to driving. Self-driving vehicles are an intelligent mechanism for drivers to pay attention to more important things other than driving. It helps invest driving time, which could amount to a lot of man-hours, in work or recreation. In addition, automated cars can reduce the usage of fuel that leads to environmental pollution.
In addition, self-driving car can aid senior citizens and the young. It can help physically-challenged individuals have enhanced mobility as they do not need to apply efforts for commuting. Self-driving car could be a panacea to a variety of problems related to the limitations of road transportation.
A robotic car provides increased travelling time and enhanced highway capacity for easier rides and reduced traffic jams. Self-driving cars will encourage road transport agencies to remove road hindrances, installed for the safety of commuters and drivers.
The United States’ Highway Capacity Manual revealed that the maximum travelling capacity of highways in the United States is 2,200 passenger vehicles per hour, per lane. This pegs the current highway usage at 5%. However, Columbia University researchers said that self-driving cars can facilitate road usage up to 278%, leading to over 8,200 passenger cars per hour per lane.
More conclusions from researches also indicated that road capacity to get to 12,000 passenger cars per hour. This improvement could significantly improve traffic and high way congestion in urban centres.
Enhanced Traffic Control
There is no gainsaying that self-driving cars can create avenue for an enhanced traffic control. Robot cars need only receive communications through electronic circuits, unlike human drivers. With these vehicles, there would not be any need for road signage, traffic police officers and insurances.
Reduction in Car Theft
With the technologies used for self-driving cars, car theft will be reduced drastically. The assumption is that automated cars will not feature steering wheel. This ensures that the demand for assuming control over robotic cars more stringent means.
Many researchers believe that automated cars are capable of reducing parking spaces and opening up spaces for better use. Robotic cars are thought to encourage the introduction of new businesses and facilitate current ones.
READ ALSO: Hunger may kill many Nigerians – Nnamdi Kanu
Possible Disadvantages of Self-Driving Car
While there are so much advantages ascribed to the use of self-driving car, there are various concerns about the ways through which the self-driving technology can disappoint.
Loss of Job
One of the concerns of critics is that self-driving car can lead to loss of the source of livelihood. Critics think that the extensive use of robotic cars will cause loss of driving-centric jobs in the transport industry. Stakeholders have also anticipated resistance from unions and professional drivers whose jobs are threatened.
Self-driving vehicles will drastically reduce the numbers of humans who drive vehicles. Amazon vision is to automated deliveries, which means no more delivery man just robot steppings out of a Self-driving vehicle to make delivery.
Loss of Privacy
There are concerns about the invasion of the individual privacy of drivers when self-driving cars are put to use.
Automated vehicles are ought to feature vehicle to vehicle (V2V) and Vehicle to Infrastructure (V2I) protocols for information sharing. These facilities can be hacked and delicate information can fall into the wrong hands.
Fear of Insecurity
With the possibility of hacking automatic cars, there are concerns that self-driving vehicles could be used as instruments for terrorist attacks. Critics believe that insecurity would further deepen with the extensive use of robotic cars.
Self-Driving Car: Recent Developments
Uber Robotic Cars
In March, it was reported that a robotic Uber taxi had an accident. The vehicle collided with another vehicle causing it to flip over. Evidences indicated that the Uber vehicle was not the cause of the accident; as the driver in the other vehicle failed to comply to traffic regulations.
Tesla Motor’s Fleet of Self-Driving Cars
On October 19, 2016, Tesla Motors launched Tesla vehicles that with self-driving capabilities in the United States. The robotic cars qualify for SAE level 5 of vehicles with self-navigating abilities to desired destinations. The hardware of the vehicles includes twelve ultrasonic sensors, surround cameras and upgraded radar with processing capabilities.
After successful testing, Tesla plans to manufacture fully enabled self-driving cars before 2017 ends.
Audi Self-Driving Car
Audi introduced their E-tron Quattro concept car at the Consumer Electronic Show (CES) in January. The vehicle, based on self-driving technology, is a totally electric car capable of covering 310 miles between charges.
Google Robotic Car
Following the successful testing of Google self-driving cars, covering over 1 million miles, Google announced in December last year that they consider spinning-off their automotive subsidiary to a company called Waymo, to record more success.
To say that the world will wait another ten years to witness the deployment of self-driving cars is a fallacy – not to mention 100 years. Self-driving cars are not futuristic automotive elements. There are fleet of cars with self-driving abilities on the roads already and several models are being developed by the day.
Google, Tesla, and several other automobile companies have projected that self-driving cars will be widely deployed by 2020. Experts project that over 10 million self-driving vehicles will ply the United States highway by the end of 2020.
At the moment, several hindrances bedevil self-driving cars. There is a need for more funds to fully develop the project. Self-driving technology is cost-intensive and various federal agencies are reluctant to fund the project.
Also, there are several legal hurdles that must be thoroughly assessed. before full implementation. Many countries do not allow cars above SAE level 2 to ply their highways and roads. Several traffic laws stipulate that drivers must have always had at least one hand on the steering wheel.