More and more organisations are talking about Motion AI as an important application area where AI is going to play a crucial role in the digital transformation. With this blog we want to explain what Motion AI is all about.
Motion AI is a subset of Edge AI, and runs typically in a wearable device. It's an AI software package that understands motion and behavior of people, animals and simply anything else that moves. Motion AI is gaining more and more momentum in the following use cases:
- User experience
- Sports and fitness
- Pets and livestock
- Tracking during transportation
Understanding human motion, human behavior and the specific case of detecting falls can be important safety features for people in work places like mines, construction sites, boats, oil platforms and road construction sites. It can also be very beneficial for lone workers such as security guards, truck drivers and forestry workers.
Understanding motion and behavior of elderly people or people with chronic diseases can be used to assess if the condition of the person is improving or deteriorating.
Understanding human motion is important when the goal is to increase user experience of new products and services. For instance, understanding human motion is important when designing smart buildings and smart access systems. New types of sensors, gestures and human motion can be used to design completely new ways of interaction between humans and machines.
Sports and fitness
The obvious use case for understanding human motion is in sports and fitness. By recognising intensity and quality of exercises during a fitness program, it can help improve motivation and results by introducing gamification and virtual trainers.
Pets and livestock
Using Motion AI with pets can help increase the health and wellbeing of pets. Is the pet active enough or does it eat or sleep too much? Using Motion AI on livestock can help understand their activity and when they are feeding and drinking.
Tracking during transportation
During transportation of valuable products, Motion AI can be used to track motion and to track if the product was exposed to shocks and potentially damaged. Motion AI can be combined with GPS in order to understand the position and the motion of the product.
Requirements and challenges of implementing Motion AI
In order to measure and analyse motion, a wearable device of some sort must be provided. The device must have motion sensors, e.g. an accelerometer or a gyro and wireless communications. Most use cases share the same requirements, the need for real time detection, the need for small, low-cost hardware and long battery time.
Streaming raw sensor data to a smartphone or to the cloud consumes too much power and will quickly drain the battery of the device. The need for real time functions and the requirement for low-cost hardware are contradictory.
Location of device
Motion AI can be integrated in existing or new wearable devices. These devices can be located in many different positions on the body:
- Head (earphones, hearing aids, helmets, glasses)
- Wrist (watch, wristband)
- Upper body (jackets, vests, collar)
- Lower body (trousers, shoes, belt)
The Swedish truck company Scania just launched the connected and intelligent vest for driver safety, the C-me vest. Imagimob has supplied its Motion AI software in that project. Read more about the Scania C-me vest here!
If you want to learn more about how to successfully succeed incorporating Motion AI in your products or services, please download our Guide here.