Many of us have either seen or driven a Tesla, no doubt noting the brand’s sleek signature aerodynamic design, large panoramic glass roofs, distinct gaming-pad steering wheel and — most of all — advanced computer systems. If you live in the San Francisco area, you might have also noticed Google’s Waymo Driver being tested for parcel delivery. In fact, autonomous vehicles are on the verge of transforming our transportation systems in the same way that the first motorcar did nearly 150 years ago.
Mechanical engineers are at the forefront of this revolution. Let’s look at some of the challenges they face and the innovations they unlock.
A Brief History
Back in the 1920s, driverless cars nicknamed “phantom autos” first popped up on select American roadways. They were radio-controlled, and though they didn’t gain much traction, they had a lot of spectators. Still, truly autonomous vehicles remained largely the stuff of science fiction.
The first fully autonomous navigation system was built into the Mercedes-Benz Eureka PROMETHEUS in the mid-1980s. Its 1,092-mile trek from Munich, Germany to Copenhagen, Denmark, and back happened entirely without human intervention. At the time, its sensor technology and route guidance systems consisted mostly of cameras, but the technology was still revolutionary. Since then, advancements in artificial intelligence, machine learning and sensor technology have paved the way for the more sophisticated autonomous vehicles we see today.
Levels of Automation
According to SAE International, there are five levels of automation: At Level 0, the driver is in complete control, while a Level-5 automation functions without any intervention from human drivers. Many vehicles sold today possess some level of automated systems, such as lane departure warnings (LPW) or automatic emergency braking (AEB). This would equate to a Level-1 automation. Although engineers at companies like Tesla are working to make more cars fully automated, experts say the first fully automated vehicle may actually be a delivery truck. In fact, Torc Robotics, a company with a presence in Austin, TX that hires our graduates, is working on expanding autonomous trucking globally.
Despite these remarkable advancements, however, autonomous vehicles still face numerous challenges.
Engineering Challenges
One major issue that automated vehicles repeatedly encounter is safety and reliability in diverse driving conditions. Self-driving cars utilize the following onboard systems to interpret the environment and determine possible hazards on their path:
- range finders
- ultrasound sensors
- cameras
- radar equipment
- light detection
- ranging (3D lidar)
Despite these technological fail-safes, autonomous vehicles are still more than twice as likely to be involved in accidents. One of the reasons is that AI cannot (yet!) make emergency split decisions, such as which is the lesser of two harms when an animal jumps out into the road while another car approaches. This requires engineers to constantly find more ways to develop better algorithms that can process sensor data in real-time.
Additionally, while traditional cars have always been at risk of being broken into, an autonomous vehicle is also in danger of being hacked into. Software vulnerabilities can cause crashes or other unpredictable behavior.
From Challenge to Innovation
Since autonomous vehicles require the most up-to-date sensor technology and computer control systems, engineers are constantly advancing technology to make driverless vehicles safer and more reliable. Advancements such as lidar and advanced radar systems provide high-resolution 3D mapping of a car’s surroundings, while AI-driven algorithms enhance decision-making processes. Companies like Tesla are continuously redeveloping their Full Self-Driving (FSD) software, in which the AI learns and improves over time.
Innovations in battery technology and energy efficiency are also extending the range and performance of autonomous vehicles. For example, Tesla’s online map feature provides drivers with detailed information on their model’s range and available charging stations, a significant improvement over gas-powered vehicles that lack such real-time information.
As driverless vehicles become more common, engineers will want to tackle another challenge: redesigning our transportation infrastructure. America’s roadways are aging, and although roads are constantly updated since the interstate system was instituted in 1956, there are still many areas with inadequate roadwork, inconsistent traffic controls and signage and a lack of sensor technology.
Innovate for the Future with an Online Mechanical Engineering Degree
Self-driving-car engineers enjoy higher salaries and bonuses, and they work with some of the most cutting-edge technology today. As autonomous vehicles become more advanced, engineers who work in this field will need to master skills in artificial intelligence, computer science and robotic sensor systems.
If you aspire to design the next autonomous vehicle, you’ll want to consider a top engineering school like the Cockrell School of Engineering at The University of Texas at Austin. Its 100% online, non-thesis Master of Science in Engineering with a concentration in Mechanical Engineering offers a robust curriculum that includes advanced courses in robotics, control systems and AI. Through hands-on projects and research opportunities, students can work on real-world problems and develop solutions that push the boundaries of autonomous vehicle technology and tackle its multi-faceted challenges.
Moreover, UT Austin’s extensive networking opportunities allow students to mingle and nurture innovative ideas with like-minded individuals.
If you’re passionate about becoming a key player in the evolution of autonomous driving, learn more here.
