ip booters

Bosch has created a sensor for future wearables to guess or “learn” almost any type of exercise pulling AI

We can have Bosch as one of the main manufacturers of household appliances, but its forays, as we have seen here on occasions, go further. In addition to autonomous cars with LIDAR or a projector for smart glasses , now they are trying it in the fitness field with a sensor with artificial intelligence for the detection of sports activities .

One of the aspects that we stop to review when we analyze quantifying bracelets or smartwatches (especially those more focused on sports) is what type of activities it records and if they do it well. Sometimes we find identical recording programs for different exercises or that any toning or strength exercise is included regardless of the muscle group that is being worked, and what this sensor shows us is the possibility of a more precise and automatic detection .

Differentiate between curl, press and whatever movement

Bosch explains that the BHI260AP (that’s what it’s called) has several functions, including navigation, orientation estimation, machine learning analytics, and activity logging . At the hardware level it includes a 32-bit microcontroller and an integrated sensor IMU that combines a three-axis accelerometer and a 3-axis gyroscope.

As we can see in the video they have shared, the chip is associated with its own app on the mobile from which the user starts monitoring, without previously specifying the exercise . Wearing the BHI260AP on the wrist, Bosch shows that it is able to distinguish the type of exercise and perform the movement count.

The BHI260AP comes standard with 15 learned activities whose registration can be customized with the user’s technique or style in order to increase precision . The promise is that the user also has a more detailed record of their activity, beyond the count: they have data on each movement.

Of course, it is not all artificial intelligence that facilitates its activity. What we also see is that the user can “teach” the chip a new activity (up to 50), so that in the future it enters into the automatic detection made by the chip.

Although the BHI260AP is associated with a mobile phone, its activity and registration do not depend on it and the components it integrates are sufficient, so that the data does not either (necessarily have to go to a cloud). And although what they show is a prototype that is far from the designs we see in smartwatches and bracelets today, the size of the chip allows it to be integrated into a wearable (perhaps with a screen if they take it further).

On this, although for now they have not confirmed that there are wearables ready for the chip, it is available for interested developers and manufacturers so that they can be designed. So we will be attentive to see what final design is with it and especially if it is as accurate as it promises.

Marie
Technology, communication and social networks. I like videogames, good series and playing all things that put me ahead.
top 10 booters