Motion State Estimation with Bandwidth Constraints and Mixed Cyber-Attacks for Unmanned Surface Vehicles: A Resilient Set-Membership Filtering Framework

Ziyang Wang, Peng Lou, Yudong Wang, Juan Li, Jia-Sheng WangSensors 2024 · DOI

TL;DR

This paper introduces a novel resilient set-membership filtering framework for Unmanned Surface Vehicles (USVs) that can accurately estimate their motion state even when facing bandwidth constraints and mixed cyber-attacks. By combining a Binary Coding Scheme (BCS) with a robust estimation approach, the framework significantly improves estimation performance and resilience in challenging real-world scenarios.

What Problem Does This Paper Address?

Unmanned Surface Vehicles (USVs) are increasingly used for tasks like marine surveillance and autonomous navigation. Accurate real-time estimation of their motion state (e.g., sway, yaw, roll velocities) is crucial for these operations. However, USVs often operate in environments with limited communication bandwidth and are vulnerable to cyber-attacks that can corrupt sensor data. Traditional state estimation methods, like Kalman filtering, struggle with these issues, especially when dealing with unknown but bounded (UBB) noises, model uncertainties, and variations in estimator gains. Existing set-membership filtering approaches have also shown limited application in USV scenarios, particularly when considering the combined challenges of cyber-attacks and bandwidth limitations. This paper aims to bridge this gap by developing a robust and resilient estimation framework that can handle these concurrent issues.

Methods

The core of this paper is a resilient set-membership filtering framework designed for USV motion state estimation. Key methodological components include:

  • Binary Coding Scheme (BCS): To address bandwidth constraints, the paper employs a BCS to encode sensor data into binary strings. This reduces the amount of data transmitted. A novel aspect is a dynamic binary bit string length selection strategy that adapts to changing network bandwidth, further optimizing communication efficiency.
  • Set-Membership Estimation: This approach is used to estimate the USV's state within a defined set, accounting for unknown but bounded (UBB) noises and parameter uncertainties.
  • Resilient/Non-Fragile Design: The framework explicitly considers and mitigates the impact of estimator gain perturbations and model parameter uncertainties, which are common in real-world systems due to factors like sensor aging and environmental effects.
  • Mixed Cyber-Attack Modeling: The paper models mixed cyber-attacks as a combination of Denial-of-Service (DoS) attacks (continuous) and bias injection attacks (stochastic, following a Bernoulli distribution).
  • Mathematical Framework: The existence of the estimator is proven using mathematical induction, and the estimator gain is derived by solving a set of Linear Matrix Inequalities (LMIs). A recursive optimization algorithm is used to achieve optimal estimation performance.

Datasets

Not discussed in the paper. The evaluation is performed using simulated data based on a USV motion model.

Key Insights and Ideas

  • BCS as an Active Defense: The paper frames the BCS not just as a bandwidth-saving technique but as an "active" measure to mitigate cyber-attack influences by encoding data before transmission.
  • Dynamic Bit String Length: The introduction of a dynamic bit string length selection strategy for BCS is a practical innovation that directly addresses the variability of network bandwidth in real-world USV operations.
  • Unified Resilience: The framework uniquely combines resilience against estimator gain perturbations and parameter uncertainties, which are often considered separately.
  • Robustness to Mixed Attacks: The proposed method demonstrates a strong ability to recover from combined DoS and bias injection attacks, which is a more realistic threat scenario.
  • Set-Membership for UBB Noise: The paper highlights the advantage of set-membership filtering in handling UBB noises, which are more representative of real-world disturbances like waves and wind than Gaussian noise.

Key Results

The paper's simulation experiments demonstrate the effectiveness of the proposed framework:

  • Improved Accuracy: In Case 1 (DoS and bias injection attacks), the Mean Squared Error (MSE) for sway, yaw, and roll velocities was significantly lower compared to a traditional set-membership estimation approach. For example, the MSE for roll velocity was reduced by approximately 93.95%.
  • Reduced Computational Load: The proposed approach also showed a slight reduction in computational load compared to the traditional method.
  • Resilience to Complex Scenarios: In Case 2, which included replay attacks, eavesdropping attacks, and dynamic bandwidth occupation, the proposed framework maintained good estimation performance, effectively resisting attacks and adapting to bandwidth changes.
  • Accurate State Estimation: Figures 5-7 and 10-12 visually confirm that the proposed method accurately tracks the real state of sway, yaw, and roll velocities, even in the presence of various disturbances and attacks, with significantly less lag and deviation than the traditional approach.

Limitations and Future Work

  • Simulation-Based: The current validation is based on simulation experiments. The authors acknowledge a gap between theory and practice and plan to calibrate the physical model with large amounts of experimental data to ensure it accurately reflects actual USV characteristics.
  • Future Research Directions:
    • Multi-source Sensor Fusion: Introduce multi-source sensor fusion technology to enhance accuracy and robustness.
  • Algorithm Optimization: Continuously adjust and optimize algorithm parameters based on field test feedback.
  • Broader Applications: Extend the BCS with dynamic bit string length strategy to other autonomous vehicles.
  • Fault Estimation: Upgrade the algorithm to estimate various types of faults in USVs.

Summary updated

Abstract

This paper investigates the motion state estimation problem of the unmanned surface vehicle (USV) steering system in wireless sensor networks based on the binary coding scheme (BCS). In response to the presence of bandwidth constraints and mixed cyber-attacks in USV communication networks, this paper proposes an improved set-membership state estimation algorithm based on BCS. This algorithm partially addresses the problem of degraded performance in USV steering motion state estimation caused by mixed cyber-attacks and bandwidth constraints. Furthermore, this paper proposes a robust resilient filtering framework considering the possible occurrence of unknown but bounded (UBB) noises, model parameter uncertainties, and estimator gain perturbations in practical scenarios. The proposed framework can accurately estimate the sway velocity, yaw velocity, and roll velocity of the USV under the concurrent presence situation of mixed cyber-attacks, communication capacity constraints, UBB noises, model parameter uncertainties, and estimator gain perturbations. This paper first utilizes mathematical induction to provide the sufficient conditions for the existence of the desired estimator, and obtains the estimator gain by solving a set of linear matrix inequalities. Then, a recursive optimization algorithm is utilized to achieve optimal estimation performance. Finally, the effectiveness of the proposed estimation algorithm is verified through a simulation experiment.
BibTeX
@inproceedings{wang2024motion,
  author = {Ziyang Wang and Peng Lou and Yudong Wang and Juan Li and Jia-Sheng Wang},
  title = {{Motion State Estimation with Bandwidth Constraints and Mixed Cyber-Attacks for Unmanned Surface Vehicles: A Resilient Set-Membership Filtering Framework}},
  booktitle = {Sensors},
  year = {2024},
  doi = {10.3390/s24216834},
}
Figure