Autonomous systems Overview
The first concept of autonomous systems in the maritime and the automotive industries was introduced in the 20th century. Since then, companies have attempted to create models of what these systems can do, with at least $50 billion invested in the automotive sector in the past five years. Tesla, for example, plans to shift from its autopilot system and introduce the first-ever self-driving technology by 2020.
The Evolution of Autonomous Unmanned Ships
- The first concept of autonomous systems in the maritime industry was introduced in the book "Ships and Shipping of Tomorrow," written by Rolf Schonknecht in the 70s. It predicted a future in which computers, onshore office buildings, would be used to captain the world's vessels.
- In the 80s, the Japanese government started a project to build an autonomous ship able to operate in high seas without the interaction of crew on-board. The ship would receive support from an onshore base. In 1988, "all the systems were simulated on a computer."
- In 2011, the Korea Research Institute of Ship & Ocean Engineering (KRISO) commenced a project to develop autonomous unmanned surface vessels (USV) for maritime surveillance and survey.
- Three years later, DNVGL launched the ReVolt, "a concept for zero-emission unmanned ships, of about 1,800 DWT, that use battery power for short voyages."
- In 2015, Rolls-Royce launched an initiative related to shipping operations, the Advanced Autonomous Waterborne Applications Initiative (AAWA), to build "both autonomous and remote operation for ship navigation, machinery, and all on-board operating systems."
- In 2017, Rolls-Royce demonstrated the world's first autonomous unmanned commercial vessel.
- In the same year, KRISO would successfully test the first autonomous ship, "Aragon II," on the real sea.
- In 2018, Rolls-Royce and AXA Corporate Solutions and Wilhelmsen and KONGSBERG, respectively joined forces to develop and commercialize autonomous systems for the maritime industry.
The Evolution of Autonomous Vehicles
- The concept of autonomous vehicles or self-driving cars in the automotive industry was first introduced in the 1939 New York World’s Fair by General Motors as part of its future vision. Using the Futurama exhibit, the company displayed an automated highway system that autonomous vehicles would rely on to transport people.
- In 1958, General Motors created the Norman Bel Geddes, a self-driving car that rode on magnetized metal spikes fixed in the roadway. It was self-driven by a gadget that steered it along the electromagnetic fields in the spikes to keep it inside its assigned lane.
- The Tsukuba Mechanical Engineering Lab in Japan made some updates to the autonomous vehicle created by GM in 1977. The self-driving vehicle was fitted with cameras and linked to a computer. This allowed it to move smoothly through the road "at 20 mph via image data processing."
- In 1988, Daimler and Mercedes-Benz joined forces to develop an autonomous van on a project called VaMoRs. By using cameras and other sensors linked to a computer on-board, the car drove unassisted at a speed of 56 mph on a highway without colliding or crushing with other objects.
- It wasn't until the Defense Advanced Research Projects Agency (DARPA) organized the self-driving cars competitions in 2004, 2005, and 2007 that a race to build safer and eco-friendly autonomous vehicles really kick-started.
- In 2014, Tesla Motors launched its Autopilot technology for its electric cars. The company claims that "the cars' cameras and ultrasonic sensors have successfully predicted collisions with up to 76% accuracy and were able to prevent them with over 90% success rate."
- Last year, Tesla Motors revealed plans to shift from its autopilot system and introduce the first-ever self-driving technology by 2020.
Outlook of Autonomous Systems and Current Applications
The Growth of Autonomous Systems In The Automotive Industry
- Driven by the demand for safe and efficient traveling options, the global autonomous car market is expected to grow at a CAGR of 18.06%, during the forecast period, 2019-2025, from $24.1 billion in 2018.
- McKinsey predicted that about 15% of cars sold by 2030 will be fully autonomous, "up to 50% electrified, and up to 10% shared." Research And Markets estimate that the industry will generate approximately $60 billion by the end of that period.
- Major automotive companies have already invested more than $50 billion in the last five years to develop autonomous vehicle technology.
- Toyota's e-Palette concept is heralding the way to a future of autonomous systems in the automotive industry. The e-Palette is "a fully-automated, next-generation battery electric vehicle" designed for a variety of use cases, including delivery van, taxi, hotel room, office, and store.
- The autonomous vehicle can also "gather traffic, route-preference, and safety data, as well as analyze usage trends to determine the demand for specific services."
The Use Of Autonomous Systems In The Retail Industry
- By 2025, more than 150,000 autonomous systems will be used in retail establishments for inventory management, material handling, security, and even floor cleaning, according to a study by ABI Research.
- The market is currently valued at $11.24 billion in 2018 and is poised to reach $23.58 billion by 2026 at a CAGR of 9.6%.
- Already, autonomous systems are used in the retail industry. Lowe has employed bots, LoweBots, at select stores to assist customers with locating products and employees with monitoring inventory levels.
- At Kroger's select stores in the US, intelligent shelf technology is used to "digitally display product and price information, simplifying a labor-intensive task."
- Ahold Delhaize autonomous system, Badger Technologies’ robot, was introduced in 2019. The system is fixed with navigation systems, numerous sensors, and cameras to enable it to navigate store paths beside customers and employees to scan floors for dangerous conditions like spills, keep stock on mispriced, missing, and misplaced inventory.
Other Current Applications
- Autonomous systems, such as robotic drones, are used in the agricultural industry to determine field deficiencies in crops, such as water and dry levels, as well as the ripeness of crops.
- In the oil industry, the systems are used to monitor gas and oil pipelines and electric transmission systems, as well as regular inspections and predicting maintenance needs.
- Baker Hughes, one of the leading oil producers in the world, applied computer perception in accessing and winnowing massive amounts of data to make better decisions on when to service its pumps.
- The oil pumping stations operated by Baker depend on mission-critical pumps that cost about $200 per piece. However, if they fail, they could cost the company millions of dollars in repair costs.
- Because of this, it needed a technology that would allow it to determine which pumps needed servicing and when to service them in order to reduce "maintenance costs and improve asset management and equipment up-time."
- With historical data on the pumps and rigs pressure, timing, temperature, vibration, and other values, the company employed computer perception to access the data and eliminate extraneous values such as large movements of the pumps in order to "detect the smaller, more meaningful vibrations coming from the valves and valve seats."
- This allowed the company to determine the number of back-up pumps required at the site and thereby make informed maintenance decisions — "actions that drove $10 million in reduced service costs."
Rutgers University (RU)
- Scientists at Rutgers University applied computers, equipped with face-and object-recognition technology, to pick up the potential lines of influence among artists and works of art.
- The reason for applying autonomous technology in the field of art history to evaluate possible influences — is that there are "practically infinite permutations of art and artists to explore" and aside from the process being cumbersome, art historians could miss significant areas of investigation.
- By applying the technology, the organization was able to make connections between "1,710 images of artworks by 66 artists spanning five centuries." For example, it was able to "detect well-known parallels—between, say, Diego Velazquez’s Portrait of Pope Innocent X and Francis Bacon’s Study After Velazquez’s Portrait of Pope Innocent—as well as similarities never detected by art historians before."