Optimal Control for Terminal Guidance of Autonomous Parafoils

Abstract

This paper deals with the development of guidance, navigation and control algorithms for a prototype of a miniature aerial delivery system capable of high-precision maneuvering and high touchdown accuracy. High accuracy enables use in precision troop resupply, sensor placement, urban warfare reconnaissance, and other similar operations. Specifically, this paper addresses the terminal phase, where uncertainties in winds cause most of the problems. The paper develops a six degree-offreedom model to adequately address dynamics and kinematics of the prototype delivery system and then reduces it to a two degrees-of-freedom model to develop a model predictive control algorithm for reference trajectory tracking during all stages. Reference trajectories are developed in the inertial coordinate frame associated with the target. The reference trajectory during terminal guidance, just prior to impact, is especially important to the final accuracy of the system. This paper explores an approach for generating reference trajectories based on the inverse dynamics in the virtual domain. The method results in efficient solution of a two-point boundary-value problem onboard the aerial delivery system allowing the trajectory to be generated at a high rate, mitigating effects of the unknown winds. This paper provides derivation of the guidance and control algorithms and present analysis through simulation.

Using direct methods for terminal guidance of autonomous aerial delivery systems

“…The first observation is that compared to trajectories presented in Figs.1 and 2 all Snowflake ADS trajectories look alike. The reason for that is that the system establishes and accurately tracks the inertial trajectory developed based on the target location and prevailing ground winds [11]. Upon exiting a deployment platform (phase 0) the Snowflake ADS steers towards a rectangular loiter area (phase 1), loiters to manage energy and estimate winds (phase 2), exits loiter area (phase 3) to enter a downwind leg as a manned aircraft would do (phase 4), and eventually performs a final turn / final approach (phase 5 and 6, respectively).…”

“…Taranenko's approach was further advanced and perfected in IDVD method, specifically designed for real-time applications [9]. The details of its application to terminal guidance of ADS can be found in [11]. For consistency, the following section briefly repeats the main results.…”

“…Also, it was suggested to be used for coordinated payload delivery [10]. Recently, it was applied to terminal guidance of ADS [11].…”

“…and equating these derivatives at the terminal points to the known BC (10) yields a system of linear algebraic equations to solve for coefficients i a and i b (for each chosen value of a single varied parameter f ) [11]. Applying Eqs.…”

Autonomous Aerial Payload Delivery System "Blizzard"

“…Unlike other systems, Snowflake ADS utilizes the inertial trajectory which allows accounting for expected winds, planning and accurately tracking the final standard-approach-pattern maneuver and landing safely into the winds. 2,3 Due to several improvements to Snowflake's guidance and control algorithm introduced during several sets of tests in 2009, the touchdown accuracy of Snowflake was eventually brought down to 10m CEP. Other novelties included networking and global reach capability.…”