Multi-UAV Collaborative Search and Attack Mission Decision-Making in Unknown Environments

Zibin Liang, Qing Li, Guodong FuSensors 2023 · DOI

TL;DR

This paper introduces the Multi-UAV Distributed Self-Organizing Cooperative Intelligence Surveillance and Combat (CISCS) strategy for coordinated reconnaissance and attack missions in unknown environments. It combines a finite-time formation controller, a Q-learning-enhanced Ant Colony Optimization (ACO) for task allocation, and an extended artificial potential field for obstacle avoidance to enable efficient and safe multi-UAV operations.

What Problem Does This Paper Address?

Coordinating multiple Unmanned Aerial Vehicles (UAVs) for complex missions like reconnaissance and attack in unknown environments presents significant challenges. Centralized control systems struggle with communication bottlenecks and single points of failure. Furthermore, achieving rapid and stable formations, efficiently allocating diverse tasks among UAVs, and ensuring collision-free navigation are critical issues. This paper aims to address these by proposing a distributed, self-organizing approach that enhances autonomy, speed, and safety in multi-UAV operations.

Methods

The CISCS strategy is built upon three core components:

  • Finite-Time Formation Controller: This controller enables multi-UAVs to rapidly achieve a mission-specific formation in a finite time. It leverages a leader-follower approach and incorporates time constraints to overcome issues of slow formation times and unstable formations often seen in traditional methods.
  • Improved Distributed Ant Colony Optimization (ACO) with Q-Learning for Task Allocation: This module addresses the complex problem of assigning multiple tasks to individual UAVs. It uses an enhanced ACO algorithm integrated with Q-Learning to generate task sequences. Novel strategies like "colony disorientation" and "search transition" are introduced to broaden the search space and prevent premature convergence of the optimization process.
  • Extended Artificial Potential Field (EAPF) for Obstacle Avoidance: To ensure safe navigation, this component extends the traditional artificial potential field method. It considers internal collisions between UAVs and provides real-time obstacle avoidance paths, enhancing the safety of the multi-UAV formation.

Datasets

Not discussed in the paper. The authors utilized MATLAB/Simulink for simulations and did not mention the use of external datasets for training or evaluation.

Key Insights and Ideas

  • Distributed Control is Key: Moving away from centralized control is crucial for robust multi-UAV operations, especially in dynamic and unknown environments where communication can be unreliable.
  • Finite-Time Formation for Speed: The finite-time formation controller significantly reduces the time required for UAVs to achieve a stable formation, which is critical for rapid mission deployment.
  • Q-Learning Enhances ACO: Integrating Q-Learning with ACO provides a powerful mechanism for adaptive task allocation, allowing the system to learn and optimize task sequencing based on environmental feedback.
  • Preventing Premature Convergence: The "colony disorientation" and "search transition" strategies are clever additions to the ACO algorithm, ensuring a more thorough exploration of the solution space and avoiding suboptimal task assignments.
  • EAPF for Enhanced Safety: The extended artificial potential field method offers a more comprehensive approach to collision avoidance, considering both external obstacles and inter-UAV collisions.

Key Results

The paper demonstrates the effectiveness of the CISCS strategy through simulations.

  • Formation Time Reduction: The formation controller significantly reduced formation time compared to methods without it. For a three-aircraft static formation, it accelerated formation by 3.5080 seconds. In a five-aircraft static formation, the improvement was 3.992 seconds.
  • Improved Formation Stability: Simulations showed that the formation controller leads to more stable formations, with reduced jitter and overshooting in UAV spacing.
  • Mission Completion Time Savings: The CISCS strategy achieved substantial reductions in mission completion time. For three UAVs, it was 52.55% faster than a single UAV, and for five UAVs, it was 58.27% faster.
  • Qualitative Improvements: The simulations visually confirmed that the CISCS strategy enables multi-UAVs to execute coordinated search and attack missions effectively, with improved formation control and collision avoidance.

Limitations and Future Work

  • Assumptions: The current strategy assumes UAVs are equipped with sufficient munitions and reconnaissance equipment to engage arbitrary targets.
  • Future Research Directions: The authors suggest future work should consider:
  • More realistic reconnaissance and combat capabilities of UAVs.
  • Incorporating dynamic constraints into the mission planning.
  • Exploring multi-dimensional, multi-platform collaborative combat, including multi-kind UAVs and ground equipment.

Summary updated

Abstract

To address the challenge of coordinated combat involving multiple UAVs in reconnaissance and search attacks, we propose the Multi-UAV Distributed Self-Organizing Cooperative Intelligence Surveillance and Combat (CISCS) strategy. This strategy employs distributed control to overcome issues associated with centralized control and communication difficulties. Additionally, it introduces a time-constrained formation controller to address the problem of unstable multi-UAV formations and lengthy formation times. Furthermore, a multi-task allocation algorithm is designed to tackle the issue of allocating multiple tasks to individual UAVs, enabling autonomous decision-making at the local level. The distributed self-organized multi-UAV cooperative reconnaissance and combat strategy consists of three main components. Firstly, a multi-UAV finite time formation controller allows for the rapid formation of a mission-specific formation in a finite period. Secondly, a multi-task goal assignment module generates a task sequence for each UAV, utilizing an improved distributed Ant Colony Optimization (ACO) algorithm based on Q-Learning. This module also incorporates a colony disorientation strategy to expand the search range and a search transition strategy to prevent premature convergence of the algorithm. Lastly, a UAV obstacle avoidance module considers internal collisions and provides real-time obstacle avoidance paths for multiple UAVs. In the first part, we propose a formation algorithm in finite time to enable the quick formation of multiple UAVs in a three-dimensional space. In the second part, an improved distributed ACO algorithm based on Q-Learning is introduced for task allocation and generation of task sequences. This module includes a colony disorientation strategy to expand the search range and a search transition strategy to avoid premature convergence. In the third part, a multi-task target assignment module is presented to generate task sequences for each UAV, considering internal collisions. This module provides real-time obstacle avoidance paths for multiple UAVs, preventing premature convergence of the algorithm. Finally, we verify the practicality and reliability of the strategy through simulations.
BibTeX
@inproceedings{liang2023multi,
  author = {Zibin Liang and Qing Li and Guodong Fu},
  title = {{Multi-UAV Collaborative Search and Attack Mission Decision-Making in Unknown Environments}},
  booktitle = {Sensors},
  year = {2023},
  doi = {10.3390/s23177398},
}
Figure