“Autonomous driving” is evolving both in the sea and on land October 26, 18:49

When you want to go out, the boat will pick you up at the island.


A robot will deliver daily necessities to the front door.


The development of autonomous driving technology that gives us a sense of such a future is currently underway in various regions.


(Hiroshima Broadcasting Station Reporter Yuki Fukushima / Sapporo Broadcasting Station Reporter Takuma Okazaki)

“If more ships can navigate automatically”

Ikino Island, a remote island in the Seto Inland Sea.

In early October, a ship was laid down in a quiet, small island harbor with a population of 14.



The total length of the ship is about 6 meters.

At first glance, it is no different from a normal small ship.

However, this ship is not an ordinary ship.

It is not humans but "AI (= artificial intelligence)" that controls it.

The ship was developed by a startup company headquartered in Osaka.



Hiroto Kimura, CEO.

Mr. Kimura used to develop AI and robots at a major IT company.



Mr. Kimura has many hobbies such as scuba diving and has many opportunities to come into contact with the sea.

The company was established in March, thinking that "the more ships that can navigate automatically, the more convenient it will be for people and goods to move."

We developed a new ship in less than half a year.

The crisis faced by the "feet" of remote islands

Demonstration experiments using a new ship have been conducted since August between two remote islands, Ikino Island and the neighboring Osakikami Island.

There are no supermarkets or hospitals on Ikino Island.

Residents have 7 flights a day, which is a town-run ferry that connects Osakikamijima, which is about 3 kilometers away.



However, the ferry's deficit last year was about 90 million yen.

The current situation is that it is supplemented by national and prefectural subsidies.



Some



island residents complain

that the ferry will disappear someday

.



In the Seto Inland Sea, there are about 100 remote islands where people live and are not connected by bridges.



With the aging of seafarers nationwide, regular routes that are being abolished due to a shortage of carriers are beginning to appear.

Eight Knot CEO Hiroto Kimura


"There are many remote islands in the Seto Inland Sea. Water transportation is indispensable for maintaining the lives of remote islands. It is necessary to support with the power of technology. I feel strongly. "

Controlled by direction of travel and speed AI

A sensor is installed at the tip of the ship to detect ships moving in and out of the sea, dust floating on the sea, and the height of waves.



Based on this information, AI = artificial intelligence operates the motor.

It is a mechanism to control the direction and speed of the ship.



In anticipation of a carbon-free society, we used electricity as a power source.

Local college students also cooperate

Along with the development of the ship, a special map was created for automatic navigation.



The company helped five students from the Hiroshima College of Maritime Technology in Osakikamijima, their hometown.

The waves in the Seto Inland Sea are calm, but the flow of tides is complicated, and there are places where it cannot pass depending on the ebb and flow of the tide.



Students carefully investigated the depth of the sea between Ikino Island and Osakikamijima and the places where fishing nets were set up, using the boats used in the school training.

It took about two weeks for the students to complete the original map.



By combining the sensor installed on the ship and the map information, AI can now search for the optimal route.

Mikiya Torii, 5th year student at Hiroshima College of Maritime Technology


"At first I was wondering if automatic operation could really be done, but I can cooperate to make a good ship so that the people of the island can live on the island with peace of mind in the future. I'm glad "

Challenges seen in the experiment

Mid-October, when the experiment was nearing the end.

The reporter (Fukushima) was also able to board the ship.

When I clicked the start button on the computer screen loaded on the ship, the ship started to move automatically.



The day was sunny and the waves were calm.



I was surprised that the ship proceeded more smoothly than I expected, avoiding the trees and debris that often flowed.

On the other hand, some issues were found in the demonstration experiment for about two months.



One of them is the "waves" that pleasure boats, fishing boats, and water skis push after passing nearby.



These waves can cause the hull to sway significantly, detecting the waves as obstacles and stopping.



And the other is to follow the maritime rule "right-hand traffic".



At sea, if ships pass each other, you will have to pass on the right side of the oncoming ship.



However, depending on obstacles such as debris and wave conditions, the AI ​​may choose a route that passes through the left side of the ship.



In the demonstration experiment, in order to give top priority to safety, we switched to manual navigation as needed.



In the future, Mr. Kimura wants to improve the learning accuracy of AI and make improvements for practical use.

Hiroto Kimura, CEO of Eight Knot


"In the area of ​​remote islands where the population is declining, it is our mission to realize an environment where we can actually maintain the same life as the last person until we are the last one, using autonomous navigation technology. I think / honestly, it's not perfect yet, it's just the first step. I want to further improve the safety that is most important for the acceptance of automatic navigation vessels in the world. "

Practical use target is 2 years later

According to the Ministry of Land, Infrastructure, Transport and Tourism, under the current law, even a ship that sails automatically must be boarded by a person with a ship license to ensure safety, and it is not possible to sail unmanned.



In accordance with these rules, the company is aiming to commercialize an automatic navigation ship two years later, carrying luggage such as groceries and daily necessities, and four years later, carrying passengers.

Possibility of "automatic delivery robot"

The development of autonomous driving technology is not limited to the sea.



In Ishikari City, Hokkaido, something strange was running on the road in September.

Approximately 2.5 meters in length.

Approximately 1.7 meters high.

Approximately 1 meter wide.

Although the width is a little narrow, it is about the size of a light wagon.



However, there is no driver's seat.


It is an "automatic delivery robot" that loads luggage and automatically runs to the destination.



From mid-August to mid-September, an information and telecommunications company in Kyoto conducted a demonstration experiment using a roadway with the cooperation of a local business operator.

Demonstration experiment on the nation's first "roadway"

Demonstration experiments of automatic delivery robots have been conducted in various places.

However, most of them are experiments that connect one place from the place where the cargo is received to the place where the cargo is delivered.



This time, it was done by collecting packages from multiple clients and transporting them to multiple destinations.

The more delivery destinations there are, the larger the size of the robot itself.

You also need to run faster to get around as many places as you can.



For this reason, robots have to deliver through the road.



According to the National Police Agency, this is the first experiment in which an automatic delivery robot carries cargo while passing through a roadway.

Easy delivery with a dedicated app

Use the smartphone app to collect and deliver your luggage.



When the store clerk receives an order for a product from a customer, the app requests the robot to deliver it.



After the robot arrives, when you operate the app again, the locker door on the side will open and you can pack your luggage inside.



There are 20 large and small lockers, which can be used according to the size of the item.

Recipients will be notified by the app when their arrival time is approaching.



When the robot arrives at the designated location, you will receive the item and the delivery will be completed.



The robot has a pre-programmed route of about 6 kilometers, which allows it to run while observing traffic rules such as pausing.



The maximum speed is 12 km / h.

Cameras and sensors are mounted on the car body to identify surrounding cars and obstacles.

Labor shortage, corona ... High expectations

In this experiment, we were monitoring at the office of the company to the last, and when we judged that it was dangerous, we switched from automatic driving to remote control.



During the experiment period, there were scenes where sudden braking and sudden deceleration were performed.

Hiroshi Yoshida, the leader of the experiment, said, "This experiment is at the stage where infants are finally able to ride a bicycle."



There are still many non-technical hurdles to put into practical use, such as the maintenance of insurance in the event of an accident.



Verification to improve safety is also indispensable.

Mr. Yoshida and his colleagues will continue to carry out demonstration experiments.

Hiroshi Yoshida, Deputy General Manager of KYOCERA Communication Systems


"We will continue to carry out several demonstrations from both the technical part of autonomous driving and the operation part of how to operate the robot, and evaluate the safety and economic efficiency. I want to do it "

While the need for delivery is increasing due to the expansion of the use of online shopping, the labor shortage of drivers responsible for delivery continues.



It is said that it is not yet clear when it will be put into practical use, but Mr. Yoshida said, "I think it will be possible to create a state in which robots are running on a daily basis faster than people think. I'm thinking, "he said, showing confidence in the early commercialization.



Development of autonomous driving technology that advances in various places.

I would like to continue to pay attention to whether it will lead to solving the problems of the existing society.

Hiroshima Broadcasting Station Reporter


Yuki Fukushima


Joined in 2021 In charge of economic coverage in Hiroshima Prefecture.

I love islands and have experience of visiting about 30 manned islands in the Seto Inland Sea.

Sapporo Broadcasting Station Reporter


Takuma Okazaki


Joined in 2017 After working at the Asahikawa station, he is currently covering the New Chitose Airport and the local economy at the Chitose branch office.