We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.SY

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Electrical Engineering and Systems Science > Systems and Control

Title: Integrated and Adaptive Guidance and Control for Endoatmospheric Missiles via Reinforcement Learning

Abstract: We apply a reinforcement meta-learning framework to optimize an integrated and adaptive guidance and flight control system for an air-to-air missile. The system is implemented as a policy that maps navigation system outputs directly to commanded rates of change for the missile's control surface deflections. The system induces intercept trajectories against a maneuvering target that satisfy control constraints on fin deflection angles, and path constraints on look angle and load. We test the optimized system in a six degrees-of-freedom simulator that includes a non-linear radome model and a strapdown seeker model, and demonstrate that the system adapts to both a large flight envelope and off-nominal flight conditions including perturbation of aerodynamic coefficient parameters and center of pressure locations, and flexible body dynamics. Moreover, we find that the system is robust to the parasitic attitude loop induced by radome refraction and imperfect seeker stabilization. We compare our system's performance to a longitudinal model of proportional navigation coupled with a three loop autopilot, and find that our system outperforms this benchmark by a large margin. Additional experiments investigate the impact of removing the recurrent layer from the policy and value function networks, performance with an infrared seeker, and flexible body dynamics.
Comments: Preprint for 2023 Scitech GN&C submission
Subjects: Systems and Control (eess.SY); Robotics (cs.RO)
DOI: 10.13140/RG.2.2.35699.45609
Cite as: arXiv:2109.03880 [eess.SY]
  (or arXiv:2109.03880v6 [eess.SY] for this version)

Submission history

From: Brian Gaudet [view email]
[v1] Wed, 8 Sep 2021 19:14:56 GMT (2723kb)
[v2] Wed, 17 Nov 2021 22:04:12 GMT (2580kb)
[v3] Fri, 31 Dec 2021 21:16:32 GMT (2861kb)
[v4] Fri, 11 Feb 2022 21:25:04 GMT (2855kb)
[v5] Mon, 18 Apr 2022 23:18:13 GMT (2922kb)
[v6] Tue, 3 May 2022 20:25:49 GMT (2918kb)

Link back to: arXiv, form interface, contact.