The feasibility criterion of fuel-optimal planetary landing using neural networks

Song, Yu; Miao, Xinyuan; Cheng, Lin; Gong, Shengping

doi:10.1016/j.ast.2021.106860

Cited by 31 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Adaptive Tanh := e nαx − e −nαx e nαx + e −nαx Adaptive ReLU := max(0, nαx) (18) In addition to these AAFs, another has been proposed as a new class of AF called rowdy [34]. The rowdy activation function allows NNs to exploit various properties of distinct AAFs and combine them into a single function by summing them, σ(x) = N k=1 σ k (αx).…”

Section: Adaptive Activation Functions For Neural Networkmentioning

confidence: 99%

Optimal Reusable Rocket Landing Guidance: A Cutting-Edge Approach Integrating Scientific Machine Learning and Enhanced Neural Networks

Çelik,

Demirezen

2024

IEEE Access

View full text Add to dashboard Cite

This study presents an innovative approach that utilizes scientific machine learning and two types of enhanced neural networks for modeling a parametric guidance algorithm within the framework of ordinary differential equations to optimize the landing phase of reusable rockets. Our approach addresses various challenges, such as reducing prediction uncertainty, minimizing the need for extensive training data, improving convergence speed, decreasing computational complexity, and enhancing prediction accuracy for unseen data. We developed two distinct enhanced neural network architectures to achieve these objectives: Adaptive (AQResNet) and Rowdy Adaptive (RAQResNet) Quadratic Residual Neural Networks. These architectures exhibited outstanding performance in our simulations. Notably, the RAQResNet model achieved a validation loss approximately 300 times lower than the standard architecture with an equal number of trainable parameters and 50 times lower than the standard architecture with twice the number of trainable parameters. Furthermore, these models require significantly less computational power, enabling real-time computation on modern flight hardware. The inference times of our proposed models were measured in approximately microseconds on a single-board computer. Additionally, we conducted an extensive Monte Carlo analysis that considers a wide range of factors, extending beyond aerodynamic uncertainty, to assess the robustness of our models. The results demonstrate the impressive adaptability of our proposed guidance policy to new conditions and distributions outside the training domain. Overall, this study makes a substantial contribution to the field of reusable rocket landing guidance and establishes a foundation for future advancements.

show abstract

Section: Adaptive Activation Functions For Neural Networkmentioning

confidence: 99%

Optimal Reusable Rocket Landing Guidance: A Cutting-Edge Approach Integrating Scientific Machine Learning and Enhanced Neural Networks

Çelik,

Demirezen

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Leveraging the continuous development and breakthrough of deep learning technology, the idea of using deep neural networks (DNN) in the guidance, navigation, and control has been widely explored in the literature [17][18][19][20]. Among all deep learning technologies, supervised learning has been extensively discussed.…”

Section: Introductionmentioning

confidence: 99%

Powered Landing Control of Reusable Rockets Based on Softmax Double DDPG

Zhang

Dong

et al. 2023

Aerospace

View full text Add to dashboard Cite

Multi-stage launch vehicles are currently the primary tool for humans to reach extraterrestrial space. The technology of recovering and reusing rockets can effectively shorten rocket launch cycles and reduce space launch costs. With the development of deep representation learning, reinforcement learning (RL) has become a robust learning framework capable of learning complex policies in high-dimensional environments. In this paper, a deep reinforcement learning-based reusable rocket landing control method is proposed. The mathematical process of reusable rocket landing is modelled by considering the aerodynamic drag, thrust, gravitational force, and Earth’s rotation during the landing process. A reward function is designed according to the rewards and penalties derived from mission accomplishment, terminal constraints, and landing performance. Based on this, the Softmax double deep deterministic policy gradient (SD3) deep RL method is applied to build a robust reusable rocket landing control method. In the constructed simulation environment, the proposed method can achieve convergent and robust control results, proving the effectiveness of the proposed method.

show abstract

“…Supervised learning techniques such as recurrent neural networks are leveraged in guidance navigation and control systems to provide solutions to optimal control problems and the analysis of orbital data files; CNNs are employed to assist the spacecraft in the landing phase through visual processing of moon images [7]; in Ref. [8], a Deep Neural Network (DNN) determines the feasibility conditions associated with the fuel-optimal powered landing; in Ref. [9], an AI-based method is developed to detect, among different features, anomalies in space objects maneuver history.…”

Section: Introductionmentioning

confidence: 99%

Real-time space object tracklet extraction from telescope survey images with machine learning

et al. 2022

View full text Add to dashboard Cite

In this study, a novel approach based on the U-Net deep neural network for image segmentation is leveraged for real-time extraction of tracklets from optical acquisitions. As in all machine learning (ML) applications, a series of steps is required for a working pipeline: dataset creation, preprocessing, training, testing, and post-processing to refine the trained network output. Online websites usually lack ready-to-use datasets; thus, an in-house application artificially generates 360 labeled images. Particularly, this software tool produces synthetic night-sky shots of transiting objects over a specified location and the corresponding labels: dual-tone pictures with black backgrounds and white tracklets. Second, both images and labels are downscaled in resolution and normalized to accelerate the training phase. To assess the network performance, a set of both synthetic and real images was inputted. After the preprocessing phase, real images were fine-tuned for vignette reduction and background brightness uniformity. Additionally, they are down-converted to eight bits. Once the network outputs labels, post-processing identifies the centroid right ascension and declination of the object. The average processing time per real image is less than 1.2 s; bright tracklets are easily detected with a mean centroid angular error of 0.25 deg in 75% of test cases with a 2 deg field-of-view telescope. These results prove that an ML-based method can be considered a valid choice when dealing with trail reconstruction, leading to acceptable accuracy for a fast image processing pipeline.

show abstract

The feasibility criterion of fuel-optimal planetary landing using neural networks

Cited by 31 publications

References 25 publications

Optimal Reusable Rocket Landing Guidance: A Cutting-Edge Approach Integrating Scientific Machine Learning and Enhanced Neural Networks

Optimal Reusable Rocket Landing Guidance: A Cutting-Edge Approach Integrating Scientific Machine Learning and Enhanced Neural Networks

Powered Landing Control of Reusable Rockets Based on Softmax Double DDPG

Real-time space object tracklet extraction from telescope survey images with machine learning

Contact Info

Product

Resources

About