Skip to main content

Gibson Innovations predicts smart audio device battery lifetime with data analytics

Within the framework of the European project DEWI, Sirris cooperated with Gibson Innovations and imec on a use case, which aimed to reduce maintenance costs and improve the customer’s ease of use of Wireless Sensor Network (WSN). A part of the project was battery-lifetime prediction.

Gibson Innovations provides music and audio products to consumers and professionals. Its focus area is on sound, design and innovation. The role of the Gibson Innovations’ Lab, situated in Leuven, is to design and develop innovative solutions based on technology trends, consumer needs, business and strategic direction, to create innovative new products and feature concepts, integrate advanced and emerging technologies into their products, define a product architecture and support product development. 

In the context of the DEWI project, Gibson Innovations’ interest was to explore and develop Smart Audio Systems that interact with various WSN devices (e.g. motion sensors) in the home, such that for example when a user arrives home, the music starts playing, the music follows the user throughout his house and the lighting is adjusted based on the context.

The collaboration between Sirris and Gibson Innovations more particularly focused on predicting the lifetime of WSN devices and smart audio devices based on their use. The possibility for Gibson Innovations to predict how long a battery of a sensor module or a smart audio device will last can help to increase customer satisfaction. For example, if a customer would be using the speakers while watching a film, the ability to predict how long a speaker will last would allow the customer to know for how long he can watch the movie without needing to charge the speakers.

Smart Home Entertainment Demonstrator

At Gibson Innovations’ premises a permanent demo-setup representing a small residential building has been installed, integrating a variety of heterogeneous wireless sensors (motion sensors, door sensors, a doorbell sensor,…), actuators and wireless audio speakers.

Gibson Innovation “Smart Home Entertainment” demo-setup

Battery-lifetime prediction

To collect data that can be used to predict a battery’s lifetime, Gibson Innovations equipped two battery-driven wireless surround speakers in this demo-setup with wireless voltage sensors, so that the battery’s state of charge could be monitored. Thanks to this setup, data on the battery’s state and playback time was collected. Part of this collected data was used to train the developed prediction models, while the rest of the data was used later to validate the models.

Three prediction models were developed, based on three different approaches, namely linear regression, Bayesian networks and polynomial curve fitting. For each of these methods, their requirements with respect to available data, model design and deployment were studied, as well as the tradeoff between model complexity and accuracy.

Sirris validated the developed prediction models on the collected data and based on the results and the requirements analysis it was decided to implement the prediction model based on the polynomial curve fitting in the "Smart Home Entertainment" demonstrator. This approach can also be used to predict the lifetime of other WSN or smart audio devices.