Predictive distance-based Send-on-Delta in trajectory tracking under network constraints

Send-on-Delta (SoD) transmission strategies takes decision to send a data-packet based on a relationship between the data, a reference data and a threshold called delta-value. This work presents a solution to determine dynamically the delta-value for the linear predicted Send-on-Delta strategy given a desired average throughput. The proposed solution is based on the determination of a relationship between the delta-value and the throughput, and modeling in real-time the transmission rate. The goal is to reduce the tracking error relative to the traditional periodic transmission methods. Simulation results show the feasibility of the method tested for continuous uniform signals and a robot trajectory in a cyber-physical space.