3D Path-Following Guidance via Nonlinear Model Predictive Control for FW-sUAS
C. Alexander Hirst, Chris Reale, Eric Frew · 2025
Read 3D Path-Following Guidance via Nonlinear Model Predictive Control for FW-sUAS by C. Alexander Hirst, Chris Reale, Eric Frew (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-24326.txt.
Introduction
Camron Alexander Hirst, Chris Reale, and Eric Frew guide a small fixed-wing aircraft along three-dimensional paths. Their method is nonlinear model predictive control. It plans control moves over a horizon while respecting the aircraft's dynamics. Two guidance algorithms are presented. Both were implemented and flight-tested on the RAAVEN uncrewed aircraft.
Control-augmented modeling and system identification come first. The first controller schedules a fixed reference path rate. That choice rewards constant inertial speed. The second controller optimizes the path rate as it flies. It trades path progress against distance from the path. Both work on general smooth curves parameterized by arc length. Flights covered several high-curvature test paths. A baseline lookahead law served as the comparison. The predictive controllers proved feasible and superior. Ground speeds reached thirty-six meters per second.
Catalog record
| CREATOR | C. Alexander Hirst, Chris Reale, Eric Frew |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 52 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-24326
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
3D Path-Following Guidance via Nonlinear Model Predictive Control for FW-sUAS C. Alexander Hirst, Chris Reale, Eric Frew Abstract This paper presents the design, implementation, and flight test results of two novel 3D path-following guidance algorithms based on nonlinear model predictive control (MPC), with specific application to fixed-wing small uncrewed aircraft systems. To enable MPC, control-augmented modelling and system identification of the RAAVEN small uncrewed aircraft is presented. Two formulations of MPC are then showcased. The first schedules a static reference path rate over the MPC horizon, incentivizing a constant inertial speed. The second, with inspiration from model predictive contouring control, dynamically optimizes for the reference path rate over the controller horizon as the system operates. This allows for a weighted tradeoff between path progression and distance from path, two competing objectives in path-following guidance. Both controllers are formulated to operate over general smooth 3D arc-length parameterized curves. The MPC guidance algorithms are flown over several high-curvature test paths, with comparison to a baseline lookahead guidance law. The results showcase the real-world feasibility and superior performance of nonlinear MPC for 3D path-following guidance at ground speeds up to 36 m / s \text{m}/\text{s} . I Introduction Fixed-wing small uncrewed aircraft systems (FW-sUAS) are aerial robotic platforms which have broad use cases in a variety of academic and commercial applications. FW-sUAS generally exhibit greater payload capacity, endurance, airspeed, and robustness compared to multirotors, desirable characteristics for many types of field missions. However, due to nonlinear dynamics, high inertia, low thrust-to-weight ratios, imprecise sensors and actuators, wind disturbances, and onboard computational constraints, high-performance path-following guidance for FW-sUAS is challenging. It follows that common FW-sUAS guidance algorithms are generally closed-form expressions designed for simple classes of paths (i.e. Dubin’s paths [ 2 ] ), and exhibit conservative performance. Furthermore, airspeed setpoints are assumed stationary and prescribed by a human operator, limiting performance in terms of rapid path progression. While modern robotic planning algorithms can generate high-utility 3D path solutions for complex missions [ 28 ] , execution of such paths remains a challenge with current guidance methods. This paper addresses a fundamental challenge in the control stack for autonomous FW-sUAS: given a smooth 3D curve (i.e. path) to follow, formulate a guidance algorithm which minimizes distance from the path while maximizing path progression. This objective has significant relevance in time-critical applications such as search-and-rescue and atmospheric science. Fig. 1 : A RAAVEN aircraft, outfitted with an array of atmospheric sensors, conducting a low-pass maneuver during a scientific field campaign. To achieve fast and accurate path following, this work leverages nonlinear model predictive control (MPC) to generate near-optimal guidance commands in real time. Nonlinear MPC enables simultaneous handling of nonlinear dynamics, state/input constraints, and non-myopic control horizons, all of which are critical to enable safe, efficient, and complex behaviors. MPC has risen to prominence as a state-of-the-art methodology for various robotic tasks, such as legged locomotion [ 29 ] , agile quadrotor flight [ 36 ] , and autonomous car racing [ 26 ] . This work builds upon prior work on MPC for FW-sUAS guidance [ 38 , 20 ] to enable high-performance 3D path-following guidance over generic curves. As a preliminary step, control-augmented modelling and system identification results for the RAAVEN FW-sUAS (Figure 1 ) are presented. Then, two novel formulations of path-following MPC for FW-sUAS are presented in detail. The first prescribes a constant reference path rate to define the MPC cost function. The second, inspired by prior works on model predictive contouring control [ 24 , 4 , 36 ] , incorporates the reference path rate as a decision variable within MPC. Finally, we deploy both MPC algorithms onboard an embedded companion computer in real-world flight tests to rigorously evaluate the feasibility and utility on a variety of 3D paths. Comparisons to a baseline lookahead controller [ 3 , 30 ] reveal the superior performance of MPC for FW-sUAS guidance. In summary, this paper makes the following contributions: 1. A control-augmented model of the RAAVEN FW-sUAS. 2. A detailed formulation of two novel 3D path-following MPC guidance algorithms for use with FW-sUAS along with descriptions of their implementations. 3. Fielded experiments to demonstrate the real-world feasibility and superior performance of these guidance algorithms on a set of challenging paths. I-A Related Work Literature on path-following methods for FW-sUAS is vast and includes geometric, vector-field, and LQR-based approaches. For more information, we refer the interested reader to surveys on 2D and 3D path-following guidance methods for FW-sUAS [ 39 , 32 ] . These approaches, while robust and computationally efficient, are conservative to ensure the vehicle remains in a safe flight envelope. As a result, these methods generally do not explicitly reason about the aircraft dynamics or limitations. For common robotic applications such as industrial arms, quadrotors, and ground vehicles, nonlinear MPC has achieved significant attention as a high-performance path-following strategy [ 12 , 10 , 13 , 16 , 41 , 21 , 36 ] . Theoretical requirements for stability and recursive feasibility were developed via terminal costs and constraints [ 12 ] and further refined for an airborne wind energy system [ 10 ] . The authors then used a truncated path-following MPC scheme (no provably stabilizing terminal costs or constraints) to demonstrate the approach onboard a robotic arm [ 13 ] . Stability was empirically demonstrated by choosing a sufficiently long MPC horizon [ 16 , 18 ] . Due to the difficulty in deriving stabilizing terms, truncated path-following MPC formulations are often applied for real-world applications onboard complex vehicles, exhibiting stability and recursive feasibility in practice [ 26 , 23 , 4 , 41 , 21 , 36 ] . This work pairs the truncated MPC approach with long horizons to demonstrate high-performance path-following MPC onboard FW-sUAS. Nonlinear model predictive control for FW-sUAS guidance has received limited attention from the research community over the last two decades. Early work on nonlinear MPC for FW-sUAS largely focused on 2D lateral control [ 22 , 42 ] with simple kinematic models, tested in simulated environments. Later works extended this idea to three-dimensional guidance, with an emphasis on increasing closed-loop performance and exploiting more of the flight envelope. It has been demonstrated that MPC could conduct fast turn and flip maneuvers, as well as obstacle avoidance [ 16 ] . Lookahead guidance laws have been used to hot-start an iterative optimization algorithm [ 15 ] . That approach enabled fast convergence of the algorithm, while maintaining a feasible backup solution in case of non-convergence. Sampling-based optimization was proposed to enable aggressive flight through a series of waypoints [ 33 ] . While effective, the algorithm was not demonstrated to be real-time feasible on an embedded computer. Recently, a different approach was taken, where MPC was used to perform path-following guidance through direct actuator control [ 35 ] . While that approach allows for lower-level integration, low-level predictive control requires detailed aerodynamic modelling which is time consuming and expensive. Furthermore, actuator-level update rates are commonly on the order of 100 Hz for FW-sUAS, a challenge for MPC on embedded hardware [ 6 ] . A shortcoming of those prior 3D MPC guidance works [ 16 , 15 , 33 , 35 ]