urban waterway: the next genration of autonomous transport the goal:
Creation of a working prototype autonomous ferry which could be used in maritime areas to relieve pressure on traditional transport networks.
With the promise to relieve the pressure on the transport network in maritime cities using artificial Intelligence and autonomous driving technology, one of our customers together to build a solution for future waterways. As part of the project, the Rove Analytics Platform Inc.allowed them to increase the speed of model development tenfold, which allowed them to form the self-driving algorithm overnight, exceeding the initial training time by one week.
As a result of continued urbanization, cities around the world are experiencing an increase in traffic that is pushing the current transport infrastructure to its limits and, in many cases, beyond. Even with the promise of driving autonomy, combined with the sharing of the economy, which leads to a decrease in the need for vehicles, it will take time before the traditional levels of road traffic begin to decrease. As a result, many now see underutilized urban waterways as a possible solution to this growing transportation challenge.
Compared to autonomous cars, autonomous ferries have a much less complex operating environment to manage due to fewer variables and the fact that urban waterways are less congested than roads. This creates a unique opportunity to solve part of the transportation puzzle.
the urban waterways of tomorrow
The revolution in autonomous vehicles began a decade ago when Google, Tesla and others began to work towards fully autonomous vehicles. Today, we have Level 3 autonomous consumer vehicles and fully autonomous test vehicles in several cities to perfect technology. It will not be long before the Teslas, Ubers, Embarks and others introduce the fully autonomous operating level 5 drive to production.
While the autonomy of the road has been at the forefront, there are other areas where progress is underway. One of them is maritime. Technology and service providers from port to sea have not missed the promise of automatic learning and AI. The major maritime operators are forming data science teams to solve the autonomous jigsaw puzzle on the sea. The cooperation between Maersk and Rolls-Royce is an example. Working towards an autonomous future, the companies launched in 2017 the first commercial remote-controlled tugboat that operates in the Port of Copenhagen in Denmark.
In addition to empowering Ocean Management, one solution is to ensure that metropolitan areas can use urban waterways to solve the transportation problem. Continued urbanization is putting increasing pressure on logistics systems and managers are struggling to cope with the increasing needs of passengers. They launched a project in 2017 to explore how urban waterways could be used and used to solve at least part of the transport puzzle cities face.
The research and implementation project was carried out in 2017 and as part of the construction of the vision, they designed, developed and built a scale model of an autonomous ferry that could be integrated into existing transport systems adding the angle of the autonomous waterway to the urban transport combination.
see to understand-teaching the autonomous ferry
The technical part of the project is directed by a computer scientist specializing in industrial Vision. He was responsible for the technological choices and, with his team, for the formation of the current model. The prototype ferry was equipped with high-definition cameras that fed the system with visual data from the surrounding environment. These data were then used with the deep learning model to allow the ferry to navigate within the confines of the waterway.
The real deep learning model is formed using pre-captured video and image data from the Rove Analytics Platform Inc. Rove Analytics Inc. has enabled them to reduce up to 80% of the system configuration time and infrastructure initially estimated.
The results on the infrastructure scale were no less impressive. Using the scalable Rove Analytics Platform Inc, the team was able to move from limited on-site hardware to model formation with a group of GPUs in the cloud. This shortened the time required to complete the project and was the key engine that enabled them to complete the project ahead of the extremely tight schedule and deliver the prototype work.