Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Dynamic Control Aware Semantic Communication Enabled Image Transmission for Lunar Landing

Fangzhou Zhao, Yao Sun, Jianglin Lan, Muhammad Ali Imran

TL;DR

The paper tackles the challenge of autonomous lunar landing under unreliable local control by coupling reinforcement-learning-based landing decisions with a dynamic, control-informed semantic communication framework (DynaSC). It introduces a patch-wise, reward-driven semantic encoder/decoder built on DynamicViT, trained with masked autoencoding and knowledge distillation to selectively transmit task-relevant image regions over a Rician fading channel, thereby reducing end-to-end transmission time and energy while preserving control accuracy. The authors provide a theoretical link between the patch prioritization parameter $\delta_t$ and the real-time reward derivative, derive end-to-end latency and energy expressions, and demonstrate, through extensive simulations on high-resolution lunar terrain data, that DynaSC outperforms JPEG and is competitive with dense SemCom baselines under both AWGN and Rician channels. The results highlight the practical potential of integrating semantic transmission with control feedback for robust, low-latency remote lunar landing, offering a pathway toward edge-enabled space missions capable of handling harsh channel conditions and stringent resource constraints.

Abstract

The primary challenge in autonomous lunar landing missions lies in the unreliable local control system, which has limited capacity to handle high-dynamic conditions, severely affecting landing precision and safety. Recent advancements in lunar satellite communication make it possible to establish a wireless link between lunar orbit satellites and the lunar lander. This enables satellites to run high-performance autonomous landing algorithms, improving landing accuracy while reducing the lander's computational and storage load. Nevertheless, traditional communication paradigms are not directly applicable due to significant temperature fluctuations on the lunar surface, intense solar radiation, and severe interference caused by lunar dust on hardware. The emerging technique of semantic communication (SemCom) offers significant advantages in robustness and resource efficiency, particularly under harsh channel conditions. In this paper, we introduce a novel SemCom framework for transmitting images from the lander to satellites operating the remote landing control system. The proposed encoder-decoder dynamically adjusts the transmission strategy based on real-time feedback from the lander's control algorithm, ensuring the accurate delivery of critical image features and enhancing control reliability. We provide a rigorous theoretical analysis of the conditions that improve the accuracy of the control algorithm and reduce end-to-end transmission time under the proposed framework. Simulation results demonstrate that our SemCom method significantly enhances autonomous landing performance compared to traditional communication methods.

Dynamic Control Aware Semantic Communication Enabled Image Transmission for Lunar Landing

TL;DR

The paper tackles the challenge of autonomous lunar landing under unreliable local control by coupling reinforcement-learning-based landing decisions with a dynamic, control-informed semantic communication framework (DynaSC). It introduces a patch-wise, reward-driven semantic encoder/decoder built on DynamicViT, trained with masked autoencoding and knowledge distillation to selectively transmit task-relevant image regions over a Rician fading channel, thereby reducing end-to-end transmission time and energy while preserving control accuracy. The authors provide a theoretical link between the patch prioritization parameter and the real-time reward derivative, derive end-to-end latency and energy expressions, and demonstrate, through extensive simulations on high-resolution lunar terrain data, that DynaSC outperforms JPEG and is competitive with dense SemCom baselines under both AWGN and Rician channels. The results highlight the practical potential of integrating semantic transmission with control feedback for robust, low-latency remote lunar landing, offering a pathway toward edge-enabled space missions capable of handling harsh channel conditions and stringent resource constraints.

Abstract

The primary challenge in autonomous lunar landing missions lies in the unreliable local control system, which has limited capacity to handle high-dynamic conditions, severely affecting landing precision and safety. Recent advancements in lunar satellite communication make it possible to establish a wireless link between lunar orbit satellites and the lunar lander. This enables satellites to run high-performance autonomous landing algorithms, improving landing accuracy while reducing the lander's computational and storage load. Nevertheless, traditional communication paradigms are not directly applicable due to significant temperature fluctuations on the lunar surface, intense solar radiation, and severe interference caused by lunar dust on hardware. The emerging technique of semantic communication (SemCom) offers significant advantages in robustness and resource efficiency, particularly under harsh channel conditions. In this paper, we introduce a novel SemCom framework for transmitting images from the lander to satellites operating the remote landing control system. The proposed encoder-decoder dynamically adjusts the transmission strategy based on real-time feedback from the lander's control algorithm, ensuring the accurate delivery of critical image features and enhancing control reliability. We provide a rigorous theoretical analysis of the conditions that improve the accuracy of the control algorithm and reduce end-to-end transmission time under the proposed framework. Simulation results demonstrate that our SemCom method significantly enhances autonomous landing performance compared to traditional communication methods.

Paper Structure

This paper contains 21 sections, 2 theorems, 40 equations, 9 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Works
  3. System Model
  4. RL-based Landing Control Model
  5. SemCom Model
  6. Landing Control-aware SemCom Transceiver
  7. Semantic Encoder
  8. Semantic Decoder
  9. Training Process
  10. Property Analysis
  11. End-to-end Transmission Time
  12. Energy Consumption
  13. The Belief Property of DynaSC
  14. Simulation Results and Discussions
  15. Simulation Setup
  16. ...and 6 more sections

Key Result

Proposition 1

Assuming that the observation error caused by reduced $\delta$ follows a normal distribution independent of time. Let $z_{t1}$ and $z_{t2}$ be the observations at times $t_1$ and $t_2$ respectively, and $\sigma_{t_1}$ and $\sigma_{t_2}$ be the variances of the observed distributions at these times. the rate of change of variance satisfies

Figures (9)

  • Figure 1: The proposed RL-based lunar landing control framework.
  • Figure 2: Dynamic sparsification ViT-based SemCom encoder and decoder network.
  • Figure 3: Comparison of DynaSC, DeiT-SC and JPEG reconstructed lunar surface images with the original images under different SNR.(Flat lunar surface)
  • Figure 4: Comparison of DynaSC, DeiT-SC and JPEG reconstructed lunar surface images with the original images under different SNR.(Crater of impact)
  • Figure 5: The average transmission and computing time under different SNRs.
  • ...and 4 more figures

Theorems & Definitions (4)

  • Proposition 1
  • proof
  • Proposition 2
  • proof