According to Wikipedia: "Sensor fusion is the combining of sensory data or data derived from sensory data from disparate sources such that the resulting information is in some sense better than would be possible when these sources were used individually."
Every day we use "sensor fusion" in our routine activities. For example, we smell smoke and then look for a grayish cloud to determine the source of the smoke, its proximity, and consequent degree of danger to us. In noisy crowds when we talk with someone, we use lip reading to enable us to understand what we don't hear clearly. Humans use sensor fusion every day to make choices based on data that is interdependent, or incomplete, versus using only one of our five senses. The better the sensor fusion, the better the choices and the more "actionable" the "situation awareness" is.
GeckoOrient automatically and intelligently merges sensor data from odometry (dead reckoning), a solid-state compass, and accelerometer-based gyroscopes (IMUs), for enhanced orientation accuracy while errand running, patrolling, or following a designated person.
Each orientation sensing system has its different strengthes and weaknesses. In the short run on uniformly smooth surfaces dead reckoning from odometry data works well. Accelerometers are also accurate in the short run and while essentially unaffected by differing surfaces, they accumulate error over a period of time. Magnetic compasses are pretty good most of the time, but can be impacted by stray magnetic fields and steel beams in a building unexpectedly. Essentially none of these sensors are sufficient when low cost is a primary design goal.
However, given GeckoSystems expertise in AI and sensor fusion, paradigms and mathematical algorithms have been developed that merge these three sometimes flawed solutions into a sensor fused result that is more accurate, repeatable, andl/or reliable than any of the three, or any two of the three. Hence GeckoSystems' mobile robot solution, GeckoOrient, is low cost and sufficient for the tasks normally expected of an indoor mobile robot.