Seoul Robotics, an AI-based perception software company, aims to turn first and last mile logistics hubs for cars and trucks into beehives, in which one sensor tower controls the movements of a fleet, like a conductor of an orchestra, and hundreds of vehicles in place.
After two years of testing its technology with BMW, the startup announced its first commercial implementation at CES with the German automaker to automate fleet logistics at its production facility in Munich, deploying technology it calls “autonomy through infrastructure.”
The vehicles controlled by Seoul Robotics’ latest product, the Level 5 Control Tower (LV5 CTRL TWR), are not autonomous themselves. All that is required of them is an automatic transmission and connectivity, according to HanBin Lee, CEO of Seoul Robotics.
A web of sensors and computers running Seoul Robotics’ 3D perception software, ‘Sensr’, is strategically placed on the infrastructure in a facility. That infrastructure then senses information about the environment around the vehicles, performs calculations, makes predictions and sends commands to the vehicles themselves, which Lee says can be done safely today without a human security operator or human in the loop.
At BMW, the LV5 CTRL TWR primarily relies on about 100 lidar sensors placed throughout the facility, but Lee said the company plans to introduce cameras and radar for sensor redundancy in the future.
Most autonomous vehicle companies are fully focused on creating self-driving vehicles, complete with all their own sensors and on-board computer power that will allow a vehicle to drive in urban environments and on highways. At least for autonomous freight transport, those companies still need a human to take over at certain points, such as navigating logistics hubs or, in the case of BMW, moving newly produced vehicles from the assembly line to vehicle distribution centers.
Autonomous transportation company TuSimple has just successfully completed its first driver-out program, driving it on an 80-mile highway from facility to facility, but the startup still requires people to manage certain operations on the ground. Waymo is building autonomous truck hubs to facilitate its transfer hub model, which includes a combination of automated and manual driving, with human drivers handling first and last mile deliveries.
LV5 CTRL TWR is not intended for use on highways. It is rather aims to fill the gaps and cut costs for OEMs, transportation companies, car rental companies and possibly even airports at the first and last mile.
“The nature of the facility is it’s a very tight parking lot, and it’s a fleet of vehicles trying to get around this small facility – someone has to orchestrate it, someone has to be the control tower and make sure the vehicles go on to the right place at the right time,” Lee told BestFitnessBands. “Even if those vehicles become autonomous one day, the Level 5 Control Tower is necessary because it is a vehicle management system. Not to mention Level 4 and Level 5 is quite far away, while this system immediately offers the advantage of being basically a robotic axis in a very limited space.”
OEMs, rental car companies and transportation companies dedicate thousands of employees to simply moving vehicles from point A to point B within their own facilities. Not only is this an unnecessary use of labor, but a lot of damage and accidents happen when people — likely locals taking part-time gigs rather than highly trained drivers — drive in crowded spaces, Lee said.
By providing information from multiple vantage points, such as behind trucks and around corners, the sensor tower should eliminate blind spots, reducing collisions and creating a more reliable process, the company said.
One of the challenges faced by companies making Vehicle-to-Everything (V2X) software is latency. Around the world, V2X controls are shared with vehicles over public 4G and 5G LTE, but because Seoul Robotics operates on private property, such as that owned and operated by BMW, it transmits information on private networks where it can ensure that bandwidth is reserved for its own use cases. In addition, the vehicles at these facilities are only allowed a maximum speed of 13 miles per hour.
An advantage of using advanced V2X to automate on private property is that there is no need to work with the government to obtain driver’s licenses and there is almost no risk of vulnerable road users succumbing to an accident, Lee said.
Another challenge that V2X companies have faced in the past, especially on public roads, is the cost associated with purchasing and installing hardware, but Lee said from a logistical standpoint the unit economy works well.
“Lidar is much cheaper today, at just $1,000 to $2,000 per sensor, and the full implementation of the system costs a few million dollars,” he said. “OEMs pay upfront for the hardware cost, so we don’t pay for hardware or installation, and once the system is installed, we basically get an installation fee and a monthly license per vehicle. Because of the money they save on labor and potential damage, their ROI can be as short as one or two years.”
Other companies are working on similar technology. In 2019, Bosch and Daimler joined forces to test automated valet parking, and Lee said there are a number of startups that have not yet announced their technology publicly, but are also bidding on the BMW gig.