Shimano developing AI-powered suspension unit
The unit would use self-learning to offer course-specific suspension and dropper post actuation
Alex Hunt
Junior Tech Writer
© Shimano/ USPTO
Back in 2020 Shimano filed for a patent with the title ‘Control device, creation method for learning model, computer program and storage medium.’
Although the title isn’t all that forthcoming with easy-to-decipher specifics, what it does suggest is that the brand is working on an AI-controlled suspension and dropper post system that uses self-learning to optimise performance.
As Shimano states, they are developing, "a control device for a human-powered vehicle including a control unit is provided. The control device controls a telescopic mechanism in accordance with output information related to controlling the telescopic mechanism that is output from a learning model in association with input information related to the travelling of the human-powered vehicle. A method for creating the learning model, the learning model, and a computer-readable storage medium are also provided."
- Read more: Whoop brings in ChatGPT for AI coaching
Part of the system looks to be a head unit for controlling the operation of the suspension and dropper post. Whether that is a phone or a dedicated computer is unclear at this stage. What the images do indicate is that courses can be saved and the suspension set-ups can be accessed from a library of courses.
© Shimano/ USPTO
The patent also indicates that courses can be saved in a library for later use
From the images relating to the patent, it looks as if the system can learn a course for either cross-country or downhill and then work out how best to optimise the suspension platform throughout that section. It does not appear to be a solely automated system, with the ability for manual inputs to better help the system deliver the support the rider is looking for.
We have seen technology like this coming into the sport of mountain biking in various guises with the likes of Fox using live valve technology and RockShox with its flight attendant system.
Tom Pidcock also used an electronic suspension system from SR Suntout on his XC race bikes for all of his XCO World Cup victories. Mondraker has also dabbled with suspension technology with its MIND-equipped bikes that use telemetry units to inform the rider how best to configure their suspension for better performance.
© Shimano/ USPTO
The rider looks to not be completely removed from the equation with some level of control
This is the first time we have seen a brand looking to combine electronic on-the-fly suspension adjustment with AI learning. The brand claims that the ultimate aim is to take the need to control a bike suspension platform away from the rider saying, "proper skill is required for the rider to appropriately control the seat post and suspension during riding."
The use of a platform like this will likely be reserved for World Cup racing whilst it's in its infancy, and potentially acting as a valuable tool for suspension mechanics to dial in set-ups quicker.
It also removes some of the human error that can be an issue in suspension set-up.
AI seems to be slowly creeping into the world of cycling from every direction. Last year Whoop announced that it would be using ChatGPT to offer an AI-powered coaching service. Recently Urtopia Fusion released an e-bike with AI functionality and even UAE Team Emirates are looking to use AI to their advantage in the coming season.
Would you be excited to jump on a bike with AI integration or do you think this could all be a step in the wrong direction for cycling? Let us know in the comments below.