Terminal Adaptive Guidance for Autonomous Hypersonic
Strike Weapons via Reinforcement Learning
Brian Gaudet
University of Arizona, 1127 E. Roger Way, Tucson Arizona, 85721
Roberto Furfaro
†
University of Arizona, 1127 E. Roger Way, Tucson Arizona, 85721
An adaptive guidance system suitable for the terminal phase trajectory of a hypersonic
strike weapon is optimized using reinforcement meta learning. The guidance system maps
observations directly to commanded bank angle, angle of attack, and sideslip angle rates. Im-
portantly, the observations are directly measurable from radar seeker outputs with minimal
processing. The optimization framework implements a shaping reward that minimizes the
line of sight rotation rate, with a terminal reward given if the agent satisfies path constraints
and meets terminal accuracy and speed criteria. We show that the guidance system can adapt
to off-nominal flight conditions including perturbation of aerodynamic coefficient parameters,
actuator failure scenarios, sensor scale factor errors, and actuator lag, while satisfying heating
rate, dynamic pressure, and load path constraints, as well as a minimum impact speed con-
straint. We demonstrate precision strike capability against a maneuvering ground target and
the ability to divert to a new target, the latter being important to maximize strike effectiveness
for a group of hypersonic strike weapons. Moreover, we demonstrate a threat evasion strategy
against interceptors with limited midcourse correction capability, where the hypersonic strike
weapon implements multiple diverts to alternate targets, with the last divert to the actual target.
Finally, we include preliminary results for an integrated guidance and control system in a six
degrees-of-freedom environment.
I. Introduction
T
ere are multiple challenges associated with designing a guidance system for the terminal phase of a hypersonic
strike weapon (HSW), where the goal is to impact a mobile and potentially maneuvering ground target. There are
multiple objectives, including precision strike capability, maximizing impact speed to enhance lethality through kinetic
energy, minimizing flight time to give the target less time to deploy defensive countermeasures, evading surface to air
missiles, and satisfying path constraints on heating rate, dynamic pressure, and load. Moreover, the guidance system will
likely be operating in a GPS denied environment where the target’s location is bounded, but uncertain. Consequently,
the HSW will need to autonomously meet mission objectives by selecting a target, potentially while coordinating with
other strike weapons and evading target defensive systems. And since the target will have likely moved from its expected
position, the guidance system needs to tolerate a reasonable range of initial heading errors. Once a target is selected, it
can be tracked by a seeker. The guidance, navigation, and control system must then map seeker angles, their rate of
change, range, closing speed, and rate gyro measurements to actuator commands. Finally, transitioning from 25 km
altitude to sea level at hypersonic speeds creates a large heating rate, significant structural load, and an extremely large
increase in dynamic pressure. This extreme aero-thermal environment can result in airframe deformation and control
surface ablation, and the vehicle will experience off-nominal flight conditions that differ substantially from that assumed
during optimization of the guidance system.
Previous work in terminal phase guidance for a hypersonic vehicle in a three degrees-of-freedom (3-DOF) environment
include [
1
], which uses optimal control theory to generate a trajectory and compares inverted to non-inverted flight
performance, with path constraints for heating rate, dynamic pressure, and load. In [
2
] the authors develop a guidance
system that allows specification of impact direction, and in [
3
] the guidance law allows specification of impact time and
angle. However, none of these papers measured performance under conditions of aerodynamic coefficient perturbation
Research Engineer, Department of Systems and Industrial Engineering, E-mail:briangaudet@arizona.edu
†
Professor, Department of Systems and Industrial Engineering, Department of Aerospace and Mechanical Engineering. E-
mail:robertof@arizona.edu
1
