Tight GNSS/INS integration as a constrained least-squares problem

In this paper, we study a new GNSS/INS tight integration algorithm. The underlying idea is to obtain user's position as the solution to a constrained version of the GNSS optimization problem, where classical trilateralization is constrained by INS measurements. This leads to a linearly constrained least-squares problem that can be efficiently solved in worst-case polynomial time. The performance of the algorithm has been validated through simulations under challenging scenarios (namely, multipath and weak signal conditions), showing an improved performance w.r.t. GNSS stand-alone and GNSS/INS Kalman filter solutions under the same conditions.