Aero Seminar: Aerospace Simulation & Guidance Systems with Kalman Filter Applications

Dr. Subhir Chaudhuri
DRDO and Associate Director
RCI

Abstract:

Classical guidance and control laws have been used for Aerospace Vehicles (which are aerodynamically unstable and non-linear) with varying degrees of performance, complexity and seeker/sensor requirements.  Accuracy requirements with dynamic tactics of modern warfare demand performance improvement, which is a trade-off between costly sophisticated hardware and computationally intensive software. Hardware-In-Loop- Simulation (HILS) facilities and methodology form a well integrated system for transforming a preliminary guidance and control system design to flight software and hardware from lift-off till end of mission. Nearly full spectrum of dynamically accurate Six-Degrees-of-Freedom (6-DOF) model can be realized. Guidance and control law studies pertaining to guidance gathering basket and stability margins are generally carried out. Hidden defects can be detected in the onboard guidance and control software as well as hardware. Model deficiencies like absence of Tail-Wag-Dog (TWD), flexibility modes and detailed seeker dynamics can be analyzed. Design modifications can be incorporated to reduce expensive flight trials. Introduction of seekers in HILS necessitates high dynamic simulators for seeker and target motion. For certain systems based on MIL-STD 1553B, HILS can be performed for validating the onboard hard real-time Guidance, Control and Navigation tasks in a distributed computing environment. This helps in Rapid Prototyping of flight systems with the state-of-the-art technology.

A target point is normally specified a priori launch for inertial guidance. Missile Guidance system needs critical Navigation states from Strap Down Inertial Navigation System (SDINS) for guiding the vehicle towards the target. Azimuth information of the missile is one of the most critical Navigation states for estimation on the moving platform before launching the missile for precise impact. It is extracted using SDINS of the missile aided by Master INS (MINS) placed on the moving platform. A 7-state Adaptive Kalman Filter (AKF) based on error states model was formulated with four measurements for finding the azimuth of the missile. This forms the core of Transfer Alignment from MINS where the gain of the innovation sequence is evaluated. Measurement residual mean and the covariance of the measurement residual are used in a unique way to compute adaptive gain based on necessity after initial accumulation of samples. A velocity matching algorithm before the launch between MINS & SDINS is used with excitation by moving platform maneuver for observing the states more appropriately. Error quaternions are used for this purpose in the feedback and the gains are selected using offline Matrix Riccati equation in discrete domain, as used in modern control systems. Estimation based filtering techniques are used in missile guidance for seekers and aided navigation systems. These techniques are validated using HILS with dynamic missile and target motion simulators.