Parameterized Trajectory Planning for Dynamic Soaring

Li, Zhenda; Langelaan, Jack W.

doi:10.2514/6.2020-0856

Cited by 7 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another work by Li and Langelaan introduced a parameterized trajectory planning method that aimed to solve the lengthy computation times of numerical trajectory optimization [26]. The deep neural network resulting from the actor-critic reinforcement learning method used to generalize the parameterization approach consisted of an actor and critic network, both of which were comprised of two fully interconnected layers, each with sixteen neurons.…”

Section: Suav Dynamic Soaringmentioning

confidence: 99%

Neuroevolutionary Control for Autonomous Soaring

Kim

Perez

2021

Aerospace

View full text Add to dashboard Cite

The energy efficiency and flight endurance of small unmanned aerial vehicles (SUAVs) can be improved through the implementation of autonomous soaring strategies. Biologically inspired flight techniques such as dynamic and thermal soaring offer significant energy savings through the exploitation of naturally occurring wind phenomena for thrustless flight. Recent interest in the application of artificial intelligence algorithms for autonomous soaring has been motivated by the pursuit of instilling generalized behavior in control systems, centered around the use of neural networks. However, the topology of such networks is usually predetermined, restricting the search space of potential solutions, while often resulting in complex neural networks that can pose implementation challenges for the limited hardware onboard small-scale autonomous vehicles. In exploring a novel method of generating neurocontrollers, this paper presents a neural network-based soaring strategy to extend flight times and advance the potential operational capability of SUAVs. In this study, the Neuroevolution of Augmenting Topologies (NEAT) algorithm is used to train efficient and effective neurocontrollers that can control a simulated aircraft along sustained dynamic and thermal soaring trajectories. The proposed approach evolves interpretable neural networks in a way that preserves simplicity while maximizing performance without requiring extensive training datasets. As a result, the combined trajectory planning and aircraft control strategy is suitable for real-time implementation on SUAV platforms.

show abstract

Section: Suav Dynamic Soaringmentioning

confidence: 99%

Neuroevolutionary Control for Autonomous Soaring

Kim

Perez

2021

Aerospace

View full text Add to dashboard Cite

show abstract

Optimization of Thrust-Augmented Dynamic Soaring

Sachs¹,

Grüter²

2022

J Optim Theory Appl

View full text Add to dashboard Cite

A novel hypothesis for how albatrosses optimize their flight physics in real-time: an extremum seeking model and control for dynamic soaring

Pokhrel¹,

Eisa²

2022

Bioinspir. Biomim.

View full text Add to dashboard Cite

The albatross optimized flight maneuver -- known as Dynamic Soaring (DS) -- is nothing but a wonder of physics, biology, and engineering. In an ideal DS cycle, this fascinating bird can travel in the desired flight direction, for free, by harvesting energy from the wind, and hence, it achieves a neutral energy cycle. This phenomenon has triggered a momentous interest among aeronautical, control and robotic engineering communities; if DS is mimicked, we have arrived at a new class of Unmanned Aerial Vehicles (UAVs) which are very energy-efficient during part (or the full) duration of their flight. However, the DS problem is highly nonlinear, under-actuated, and dependent on the wind profiles. This has resulted in decades of DS control literature that, while making progress in addressing the control problem, seem not to be aligned well with the nature of the DS phenomenon itself. The control works associated with DS in the literature rely heavily on constrained optimal control algorithms, control designs that require a mathematical expression of the objective function, and predefined wind profile models. Clearly, a functioning controller for DS that allows meaningful bio-mimicry of the albatross, needs to be autonomous, real-time, stable, and capable of tolerating the absence of the expression of the objective function (similar to what the bird does). The qualifications of such controller are the very same characteristics of Extremum Seeking Control (ESC) systems. In this paper, we show that ESC systems existing in control literature for decades are a natural characterization of the DS problem. We provide the DS problem setup, design, stability, and simulation results of the introduced ESC systems. The results, supported by comparison with optimal control solvers, emphasize that the DS phenomenon is a natural expression of ESC systems in nature and that DS can be performed autonomously in real-time.

show abstract

Parameterized Trajectory Planning for Dynamic Soaring

Cited by 7 publications

References 12 publications

Neuroevolutionary Control for Autonomous Soaring

Neuroevolutionary Control for Autonomous Soaring

Optimization of Thrust-Augmented Dynamic Soaring

A novel hypothesis for how albatrosses optimize their flight physics in real-time: an extremum seeking model and control for dynamic soaring

Contact Info

Product

Resources

About