Kawasaki Heavy Industries unveils shipbuilding robot for first time, plans practical deployment in 2028

news.tv-asahi.co.jp reported Oct. 6 that Kawasaki Heavy Industries had unveiled for the first time a shipbuilding robot under development. The robot is designed to move along a ship’s side on four legs and perform tasks such as welding and painting, and is said to be equipped with physical AI. Kawasaki Heavy Industries plans to conduct a demonstration next year at its Sakaide Works in Kagawa Prefecture, with practical deployment planned for 2028.

The report said Kawasaki Heavy Industries aims to bring a robot that combines mobility and work planning into processes where people are still often needed, including work at heights and on curved surfaces of ships. This makes the implementation of labor-saving technology at shipyards a concrete challenge. With practical deployment planned for 2028 and a demonstration at Sakaide Works next year, the focus is expected to shift from the concept to its suitability for real-world worksites.

What is new in this report is that Kawasaki Heavy Industries showed an actual robot for shipbuilding that can move along a ship’s side on four legs while performing welding and painting. Rather than simply introducing an industrial robot as a standalone unit, the approach combines the robot’s mobility system with work planning to suit conditions such as working at heights and on curved surfaces. The emphasis on physical AI suggests the main aim is not only control software and perception, but also a concept for autonomously developing the sequence of work at the site. Compared with our previous coverage, Kawasaki Heavy Industries opens Physical AI center in San Jose, the company established a base for social implementation in San Jose and has now brought the technology to a specific worksite in shipbuilding. The policy of expanding beyond the scope of industrial robots, outlined in Kawasaki Heavy Industries explains its path to becoming a comprehensive robotics maker, is also now connected to a heavy-industry application involving work on ship hulls. However, the article does not make clear how far the demonstration conditions, scope of target processes and necessary safety requirements have been defined. See the original report for details. From an industry-structure perspective, automating shipbuilding involves more than installing labor-saving equipment. The key question is whether hull shapes, scaffolding conditions, the sequence of painting and welding, and work plans for each site can be designed together. The robot is significant in that it presents a mobility system and planning generation as an integrated package. If demonstrations at Sakaide Works advance toward practical deployment in 2028, the question will be whether the robot can handle processes that fixed equipment has not been able to cover. At the same time, production volumes, the division of processes for deployment, maintenance arrangements and methods for accumulating training data remain unclear. The next issue is how much repeatability can be confirmed through the demonstration.

Shipbuilding is one of Japan’s heavy industries with particularly demanding process conditions. Whether a mobile robot adapted to ship hull shapes can approach practical use is therefore an important issue for Japanese shipbuilders and suppliers of robot-control and sensor-related components. In particular, linking welding and painting to autonomous planning will require not only the robot itself but also the development of site data, making the quality of domestic demonstration planning a potential factor in competitiveness.