Optimal measurement selection algorithm and estimator for ultra-wideband symmetric ranging localization
This work addresses localization challenges for autonomous systems like drones in environments where only single range measurements are available, representing an incremental improvement over existing methods.
The paper tackles the problem of localization for rigid-body agents using single range measurements by deriving a sensor-agnostic state estimator and a greedy optimization algorithm for measurement selection, demonstrating in indoor multicopter experiments that this approach improves over naive fixed-sequence strategies and can be used for feedback control.
A state estimator is derived for an agent with the ability to measure single ranges to fixed points in its environment, and equipped with an accelerometer and a rate gyroscope. The state estimator makes no agent-specific assumptions, and can be immediately applied to any rigid body with these sensors. Also, the state estimator doesn't use any trilateration-based method to calculate position from range measurements. As the considered system can only make a single range measurement at a time, we present a greedy optimization algorithm for selecting the best measurement. Experiments in an indoor testbed using an externally controlled multicopter demonstrate the efficacy of the algorithm, specifically showing an improvement over a naïve strategy of a fixed sequence of measurements. In separate experiments, the algorithm is also used in feedback control, to control the position of the multicopter.