Making Magnetics Purposeful
Over the last two decades, fueled by the development of autonomous navigation technologies and integration of sensors in our daily life via smartphones, a plethora of novel sensing technologies have been developed and co-integrated to bring smartness to the objects we interact with. Once restricted to very specific use cases and controlled environments, technologies such as IMUs, LIDAR and GPR have democratized their use — providing novel ways to interact with our environment. LIDAR now allows us to quickly measure and digitize our world, IMUs translate our movements into an object to human interface and GPR uncovers hidden objects alike with the promise to even enable navigation. Magnetometers have the untapped potential to see beyond these sensing modalities and operate in all conditions. In this post, we dive into why getting there involves more than improving on sensitivity which drives much of the magnetometer discussions.
Magnetometers are notably missing from the recent sensor fusion excitement and sensing revolution. Despite impressive sensitivity figures, up to 1 ppm of Earth’s field and their use everyday to let your navigation system know how you are oriented or headed, their usage is still limited to very specific usecases such as submarine detection, minerals exploration, archeology and locating unexploded ordnances (UXOs). For navigation specifically, conventional magnetometers provide a base knowledge to assist other sensors such as accelerometers, gyroscopes or even base odometry. Through our customer discovery process, we discovered that end users often overlook magnetic data (if not disabling the sensors altogether) as they are too easily corrupted by their environment. At SBQ, we are shifting away from the sensitivity metric to gain a holistic view of the magnetometers relevance for end users. We are solving this sensing core weakness to focus on i) building a high precision magnetometer system that will enable the use of advanced interpretation algorithms and ii) tailoring a combination of sensors to cancel out the environment and deployment platform noise. These new capabilities, which include into a Magnetic Intelligence (MI) system, will shift magnetometers from a ‘nice to have’ sensor to a ‘must have’, ultimately enhancing the navigation systems accuracy.
To get there, our team is embracing one of our core value: collaboration. We are going to the field with end users, co-developing minimally viable products to gather user feedback and plan subsequent product iterations. This allows us to refine the core features brought by MI while integrating the new capabilities within existing end users flows. By getting closer to users workflows, the dialogue then goes into unexpected directions, often simplifying the list of features which need to be implemented. Combined with field deployment considerations, such as noise sources, the ergonomics or SWaP of the product and long term stability, all of a sudden the key technical requirements change drastically. Most importantly, this means that the magnetometer sensitivity is no longer the core metric to consider.
For example, if you are outside with a very sensitive magnetometer, what happens if the daily fluctuations (caused by solar storms) or passing by cars create a magnetic field overwhelming your sensors? If your magnetometer drifts so much that a recalibration is required every day? If by integrating your sensor onto a small robotic vehicle, drone or satellite creates noise sources so large or sensor offsets which make the magnetometer readings irrelevant?
Answering these questions is at the core of the development of SBQ’s MI system. Specifically, through the MagQuest challenge organized by the US National Geospatial-Intelligence Agency, we have been directly confronted with these questions and has forced us to adapt our whole data collection strategy to include alternative sensors and adapt the algorithms to take a holistic view of the platform. Suddenly, the diamond quantum magnetometer ability to provide the vector magnetic field without drift, heading errors or dead zones (which are currently up to 1000x larger than the sensitivity of the devices!) become key advantages to meet the target performance metrics of the mission.
This is only one of the use cases we are looking at SBQ to make magnetics more purposeful to end users. By collaborating closely with end users and a various set of expertise in our team, the Magnetic Intelligence toolkit can open new market opportunities as the enhanced capabilities solve an ever growing set of applications!
