1-Bit processing based model predictive control for fractionated satellite missions

The model predictive control (MPC) has great advantages in dealing with complex control constraints. However, traditional MPCs are too complex to implement in real-time embedded systems. This is especially true for nano-satellites due to limited on-board resources. This paper introduces a novel 1-bit processing based MPC (OBMPC) algorithm for a fractionated satellite mission, which can significantly reduce online calculations by removing multiply operations. The resulted pulse signals can be used to drive the actuator directly. The quantized state feedback fits the OBMPC in the frame work of quantized MPC. The stability issues and the design criterion are discussed in this paper. The simulation is based on a 2U CubeSat model in a fractionated satellite structure, in which the payload and actuators are separated from the controller and controlled via wireless inter-satellite link (ISL). Compared to the equivalent traditional MPC controller, FPGA implementation based performance analysis shows that OBMPC is feasible for fractionated satellite missions.