A month ago, the first self-driving bus operations begun its three-month trial in the southern Taiwanese city of Tainan. It is the first revenue-generating service of autonomous rapid transit (ART) project supported by the central and local governments. The goal is to launch commercial operation across the country in 2021.
“Autonomous technology will lead a revolutionary change to the city’s transportation systems,” said Huang Wei-Cher, Mayor of Tainan. “The smart transportation initiative will help us improve overall road safety, operational efficiency, and rural area transportation services.”
The service will cover two business districts. One service will run on weekends only, on a 2.5km route between Nanke Railway Station and National Museum of Prehistory, while the other will be on weekdays in the 6.4km route along Shalun Smart Green Energy Science City, where a smart vehicle testing site is located.
The ART project is part of Taiwan’s two-year smart transportation development plan. It is supported by the Ministry of Science and Technology, the Ministry of Transportation and Communications and the National Development Council. Tainan has been tapped as the country's first municipality to offer commercial autonomous bus services.
“Through continuous safety validation of self-driving technology, ART will be a cost-effective alternative to traditional mass rapid transit (MRT) systems for cities like Tainan,” said Kung Ming-Hsin, minister of the National Development Council. “From a public infrastructure perspective, this new approach can not only reduce deployment costs by half but also serve twice as many citizens. The pioneering innovation of ART will provide a safer, more efficient, and more accessible transportation network and become the pride of Taiwan.”
During the trial period, the bus will not carry passengers but will have someone in the driver's seat in case of any problems. The Tainan city government aims to begin passenger services on trial basis in the fourth quarter of the year and outsource the operation to the private sector in 2021, creating the first commercial driverless bus service in the country.
A look under the hood
The electric autonomous bus is equipped with six cameras and four lidars -- a remote sensing system that measures distances by illuminating a target using laser lights -- to monitor traffic conditions. It also includes high-definition maps and 5G-enabled connected car technologies.
The solution is put together by Silicon Valley-based LILEE Systems in collaboration with local partners such as Green Transit Company, Taiwan Optical Platform, HYA Company, H.P.B. Optoelectronics, Iscom Online International Information, STARTRII CO and FRED. LILEE Systems is technology systems integrator specialising in advanced wireless communications and autonomous driving solutions.
LILEE Systems used its expertise in rail systems to design and develop the autonomous self-driving bus, which is made from a repurposed diesel-run vehicle. Its ART concept is a fixed-route, fixed-speed system, similar to metro or light rail systems. Because its routes and stops are pre-defined and reserved, ART shares the same benefits as rail systems to provide scheduled and predictable services regardless of traffic conditions.
Topped with vehicle-to-infrastructure (V2I) communications and smart signals, ART further guarantees travel efficiency. The Tainan self-driving buses on virtual tracks are centrally managed and controlled from a cloud-based operations control centre(OCC) as added layers of safety.
“Thanks to the support from the governments and our partners, we are glad to bring a conceptual technology to a Level-4 autonomous bus solution,” said Jia-Ru Li, CEO of LILEE Systems. “We believe ART will become the new mode of public transportation in the coming years, solving urban transportation challenges such as driver shortages and increasing operational costs.”
Proof of concept test
In 2018, the Taiwanese government started to test the ART concept through proof-of-service trials. Thousands of people experienced a 9-meter autonomous bus on a fixed bus route in Taichung, the second largest city in Taiwan, with a speed of up to 30 km/h.
However, they had to jump passed the hurdle that met an earlier business case for autonomous cars – the technology has yet to demonstrate driverless cars’ ability to respond to a real-world traffic challenge. Technology giants and automotive industry continue to promise solutions, and the availability target is moving further and further into the future. At the beginning, unmanned personal cars on the road would likely do more harm than good to the already worsening urban traffic.
LILEE Systems’ ART concept comes from the railway industry and is based upon rail-safety principles. Autonomous buses run on a virtual track, monitored by a centralised operational control centre in real time with a fail-safe system. That means, each action taken by the autonomous bus has to be confirmed to be safe locally and remotely before it can be performed. This added layer of safety that has been practiced by the railroads for many years and LILEE Systems ported it into autonomous driving to meet the highest safety standards.
Although it is similar to a metro system, ART’s virtual tracks are based on dedicated bus lanes, meaning it does not require costly and time-consuming infrastructure development as for a rail system. Essentially, ART shares the same benefits of a metro system yet with significantly lower capital costs and faster speed to market.
Another hurdle that needed to be met is the high cost of building an autonomous vehicle from scratch. When required sensors and autonomous driving systems are added to a new $30,000 car, LILEE Systems estimated the final price can easily reach $130,000. Purchasing an autonomous car might not be realistic until required technology cost can be lower.
In the end, city governments and the public still struggle to justify the real benefits of driverless cars, resulting in the delay of adoption.
Two years ago, the POC trial in Taiwan demonstrated the feasibility of modifying regular diesel buses for ART. Without having to purchase new vehicles, it eases the financial stress of city governments and bus operators and increase their willingness to join efforts to participate in ART projects.
An $100,000 technology investment in deploying autonomous systems for a $300,000+ commercial vehicle that has the potential to serve 30+ riders each trip is way more practical than the same investment for a $30,000 compact personal car.
Senior executives from LILEE Systems believe that close government-industry collaborations are key to the success of ART. Opening new roads for testing and providing clear regulations are especially critical for continuous validations of the business model.