Immunity-Based Framework for Autonomous Flight in GNSS-Denied Environment

Abstract

A framework based on the artificial immune system (AIS) paradigm is proposed in this paper for correcting position and velocity estimations for autonomous flight vehicles in environments where global navigation satellite systems (GNSS) are not available. The AIS consists of sets of memory cells built under normal conditions when all sensor systems function properly. The memory cells mimic the functionality of memory T-cells and B-cell capable of encoding and storing information about the invading antigens and the needed antibodies. This information is used to enhance the response of the innate immune system with an adaptive component that is expected to accelerate and intensify the immune response when subsequent infections with the same antigen are experienced. The artificial memory cells are constructed with two parts. One represents the antigen and is a collection of instantaneous measurements of relevant features that characterize the dynamics of the system and are the basis of the position and velocity estimation. The other represents the antibodies and is a set of instantaneous estimation errors that are viewed as necessary corrections for the estimation. During GNSS-denied operation, the current measured features are matched against the AIS antigens and the corresponding corrections are extracted and used to tune the outputs of the position and velocity estimation algorithm for feedback control. The functionality of the proposed methodology and its promising potential is successfully illustrated using the West Virginia University unmanned aerial system simulation environment.