Driverless Automobiles: A History of Innovation - Essay Sample

Introduction

The first advances in radio and aviation technology laid the foundation for the development of driverless automobiles. The idea of self-driving vehicles in the United States started in the 1920s. Such advancements occurred due to the need to address the rising number of road accidents that resulted from careless driving (Kroger, 2016). Notably, the notion of autopilot originated in the gyroscopic Airplane Stabilizer, which could keep the airplanes stable, allowing the pilot to breathe, but not for long. Besides, radio technology offered a new dimension of using radio waves to control equipment remotely. The radio technology was primarily vital to the United States military, which channeled experiments on remote-controlled ships, aircraft, and torpedoes.

Is your time best spent reading someone else’s essay? Get a 100% original essay FROM A CERTIFIED WRITER!

Save 25%

The Radio Air Service engineers developed the first driverless automobiles in 1921 at the air force test base at McCook. The 2.5-meter automobile was controlled through radio waves from an army truck that was 30 meters behind the car. Americans, in 1925, witnessed the developments of another driverless vehicle, named the American Wonder (Kroger, 2016). Houdina Radio Control Company was responsible for the creation of the vehicle. However, the working of American Wonder exhibits that it was miles away from autonomy because while it was driverless, it was not self-driving. A driver from another car would control the American Wonder.

By the 1930s, the technology state in the United States had advanced. Resultantly, the development of driverless cars was not only limited to radio technology (Kroger, 2016). Instead, the developers would use the automated transport system that had been pioneered by the American auto and oil industries. The automated highway system also supported the dreams of establishing computerized cars. General Motors (GM) has played an instrumental role in the progress towards the development of autonomous vehicles. During the 1960 World's Fair in New York, GM staged the most memorable presentation. The Futurama, which depicted the future state of transport systems, with the possibility of achieving fully automated cars.

While GM's Futurama show had provided glimpses and imaginations of car autonomy, the Second World War crippled the United States' effort to realize the dream. However, the Second World War had led to various inventions that could be used in the establishment of driverless cars in the periods after 1940. For instance, the radar technology, which was invented during the Second World War, was vital to maintaining the distance between a driverless automobile and the car ahead of it. The magnetic detectors, which had also been discovered in the Second World War, would be essential in realizing automatic driving. In 1953, GM began testing the impression of an electronic road, in conjunction with the Radio Company of America (RCA). Autonomous driving became famous following the premiere of the Firebird II concept automobile, which occurred in the Motorama, a traveling promotional show (Kroger, 2016). In 1956, the United States built the Interstate Highway System that played an immense role in the progress towards car autonomy.

However, the efforts towards automation realized real rewards in 1958 when GM developed an automatically guided car that completed a one-mile test drive at the Company's Warren Technical Centre in Michigan. In the development, GM had fitted two electronic sensors on a 1958 Chevrolet model (Kroger, 2016). The two sensors tracked a wire that had been laid on the test route. By following the wire, the sensors controlled the vehicle's steering wheel, guiding its turn to follow the wire. In 1958, Chrysler launched the "supergadget," an autopilot feature that was sold at 86 dollars. The development of Chrysler's "supergadget" laid the real foundation of achieving autonomy in vehicles.

In the 1970s, the developments towards autonomous cars started to shift, shunning away from depending on the automatic highways. By the 1970s, Japan was also showing significant efforts in creating self-driving cars (Kroger, 2016). In 1977, a Mechanical Engineering Laboratory team from Tsukuba, Japan, under the leadership of Sadayuki Tsugawa, launched the first car that could use two cameras to capture and digest the images of the lateral rail guides on the roads. It would move at a speed of 10 km/hr. However, it lacked the lane-switching function. In 1981, Hans Moravec deployed robotic navigation systems to Stanford Cart. The developer also applied TV cameras on the cart, which was able to move autonomously in a chair-filled room, without human intervention. While the development was a significant boost to the efforts towards car autonomy, the cart was slow and would take substantial time processing information before navigating.

The 1980s witnessed vigorous studies and tests by research, organizational, and academic institutions across the globe. Ernst Dickmanns, a University of the Federal Armed Forces in Munich student, was the first person to design a driverless vehicle that could use digital processors in 1984. He used a 5-ton van, christened VaMoRs (Versuchsfahrzeug fur autonome Mobilitat und Rechnersehen), which could carry the cameras and computer systems, considering their big sizes then (Kroger, 2016). In 1987, the VaMoRs drove autonomously, depending only on the cameras, for 2o kilometers at a speed of 60 mph. The VaMoRs success increased the attention of Daimler-Benz AG towards Dickmanns' research.

However, the real efforts towards car autonomy started in the 2010s. The first electric vehicle came to the global stage in 2010 at the collaborative GM & SAIC exhibition area at the Shanghai Expo of 2010. GM's EN-V (General Motor's Electric Networked Vehicle) had technical features such as collision avoidance, vehicle platoons, as well as self-parking (Bimbraw, 2015). The 2010s saw the incorporation of the most sophisticated technology in the autonomy of the vehicles. For instance, the Audi TTS, which also premiered in 2010, deployed algorithms, emerging software, as well as electronics. Since the 2010s, autonomy has used the latest advancements in technology in the navigation systems of cars. '

Noteworthy, the current phase of autonomous vehicles deploys the latest technology. However, most of the automobiles have not attained full autonomy. Currently, the techniques used in car autonomy include algorithms, vehicular edge subsystem, as well as the cloud framework (Liu et al., 2019). The algorithm technologies include sensing, perception-localization, perception-object tracking and recognition, and decision techniques. Automated cars use sensor data to dictate the motion decisions. The information comes from multiple sensors, with each having certain advantages and setbacks. In the effort to avert the delays from each type of sensor, the cars integrate the working of multiple sensors like the Inertial Measurement Unit and the Global Navigation Satellite System (GNSS/IMU).

Light Detection and Ranging (LiDAR) is another technology that helps in localization, obstacle avoidance, and mapping. Self-driven cars also use camera technology to detect and track objects on the road. The cameras are instrumental in lane, pedestrian, and traffic light detections. For increased safety of the cars, the developers mount eight or more cameras around the vehicles (Liu et al., 2019). Radar and Sonar technologies are also crucial in the presentation of sensory data. The sonar and radar systems display the distance between the automobiles and the nearest obstacle in front. The presence of an object prompts the vehicle to either brake or turn to avoid the obstacles.

Autonomous cars also use perception technologies to interpret the sensory information to comprehend the environment of the vehicles. The Kalman filtering systems help the cars integrate all the sensory data to localize objects and obstacles around the vehicles. For instance, the particle filter information by LiDAR relies on shape description of the surrounding, making it hardly possible to differentiate the individual points (Caesar et al., 2020). The filtering technique helps contextualize the shapes to the HD maps from the cameras to achieve certainty. Self-driven automobiles also use object tracking and recognition perceptions. Automated cars use the Convolution Neural Network to accomplish the task of obstacle recognition and tracking. The CNN evaluation course occurs in four layers. The convolution layer is the first stratum, which uses various filters to extract varied characteristics from the input picture. Second, the activation layer dictates the decision to activate the specified neuron (Liu et al., 2019). Third, the pooling stratum reduces spatial representation size to minimize the parameter number and resultantly, the network computation. Lastly, the fully connected level links all neurons to all executions in the previous layer.

Autonomous cars also use real-time operating technologies. Real-time operating systems (RTOSs) seek to serve real-time frameworks such as aviation, automotive, and industrial applications. In the automotive industry, QNX is the most popular RTOS, and its kernel entails inter-process communication, timers, CPU scheduling, as well as interrupt redirection. The autonomous vehicle technologies vary among the specific car models despite serving similar or related purposes. Waymo Driver is one of the latest automated cars, whose development to the fifth phase of self-driven cars occurs from notable improvements (Lavars 2020). The car has upgraded its LiDAR sensors to the extent of detecting objects, including traffic lights and signs.

Tesla Autopilot is another autonomous vehicle. However, the automated features of the automobile have been subjected to immense criticism due to the fatalities that have occurred from its use. However, the current state of Tesla's automation exhibits the company's efforts and limitations towards automated driving. Nassi et al. (2020) argue that the Tesla technologies, including the algorithms and decision techs, are capable of detecting objects in the path of the vehicle. However, unlike the other cars, the Tesla model has already been tested and even sold, explaining the numerous complaints due to the accidents resulting from its uses. The lane-switching technologies are absent in other autonomous car models. For instance, General Motors Super Cruise models also come with automated feature technologies, which mirror the universal ones for autonomous cars, as stated earlier. Akin to Tesla's model, the General Motors self-driving cars possess sensors, cameras, and radars that help the vehicle detect objects on the road (Wayland, 2020). Contrarily, the General Motors models are yet to incorporate the technologies that prompt the cars to either switch lanes or swerve in the event of an obstacle in their paths.

Besides, the Mercedes-Benz Distronic Plus has a semi-autonomous feature, which allows the vehicle to self-drive and be driven by a human driver. The car possesses two short-range sensors as well as a long-range radar system, which help the car to achieve Adaptive Cruise Control (ACC) (Revell et al., 2020). ACC is instrumental in the event of another vehicle being in front of the Mercedes-Benz Distronic Plus. A comparison of the latest car models exhibits that only the Tesla model has achieved full autonomy, despite failing several times. Among the stated cars, only Tesla has faced the challenges of autonomy. While Waymo seems to offer better results in self-driving, it is yet to be tested. However, there are multiple chances that the model will be the best self-driving car because of the upgraded technology behind its working (Lavars, 202...