Collision Avoidance Strategies for Unmanned Aerial Vehicles in Formation Flight

Seo, Jungmin; Kim, Youdan; Kim, Taegyun; Tsourdos, Antonios

doi:10.1109/taes.2017.2714898

Cited by 130 publications

(73 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Besides, the actions the UAV uses to avoid collision include both heading angle change and velocity change. The work in [34] investigated strategies for multiple UAVs to avoid collision with moving obstacles, which is a little similar with collision with enemy UAV. However, their work assumes all UAVs and all obstacles have constant ground speeds, and the direction of the velocity vector of an obstacle is constant.…”

Section: Literature Reviewmentioning

confidence: 99%

Automated Enemy Avoidance of Unmanned Aerial Vehicles Based on Reinforcement Learning

2019

View full text Add to dashboard Cite

This paper focuses on one of the collision avoidance scenarios for unmanned aerial vehicles (UAVs), where the UAV needs to avoid collision with the enemy UAV during its flying path to the goal point. Such a type of problem is defined as the enemy avoidance problem in this paper. To deal with this problem, a learning based framework is proposed. Under this framework, the enemy avoidance problem is formulated as a Markov Decision Process (MDP), and the maneuver policies for the UAV are learned based on a temporal-difference reinforcement learning method called Sarsa. To handle the enemy avoidance problem in continuous state space, the Cerebellar Model Arithmetic Computer (CMAC) function approximation technique is embodied in the proposed framework. Furthermore, a hardware-in-the-loop (HITL) simulation environment is established. Simulation results show that the UAV agent can learn a satisfying policy under the proposed framework. Comparing with the random policy and the fixed-rule policy, the learned policy can achieve a far higher possibility in reaching the goal point without colliding with the enemy UAV.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Automated Enemy Avoidance of Unmanned Aerial Vehicles Based on Reinforcement Learning

2019

View full text Add to dashboard Cite

show abstract

“…The European Commission's Directorate General for Mobility and Transport (DG MOVE), the European Defence Agency (EDA), the European Aviation Safety Agency (EASA), and the Single European Sky Air Traffic Management (ATM) Research (SESAR) Joint Undertaking (SJU) are also stepping up the efforts to safely accommodate UAS into the European aviation and ATM system [6]. In parallel with these government and industry-led initiatives, the aerospace research community has been continuously working on several challenges of integrating UAS into non-segregated airspaces including separation thresholds and methods for small UAS [7,8], UAS encounter modelling and collision avoidance [9][10][11], 3D obstacle avoidance strategies for UAS [12][13][14][15] dynamic model augmentation [16], Global Navigation Satellite Systems (GNSS) integrity augmentation for UAS [17], surveillance sensor integration in the UAS platform [18,19] and well-clear boundary models for UAS DAA [20][21][22][23][24].…”

Section: Introductionmentioning

confidence: 99%

Probabilistic Safety Assessment for UAS Separation Assurance and Collision Avoidance Systems

2019

View full text Add to dashboard Cite

The airworthiness certification of aerospace cyber-physical systems traditionally relies on the probabilistic safety assessment as a standard engineering methodology to quantify the potential risks associated with faults in system components. This paper presents and discusses the probabilistic safety assessment of detect and avoid (DAA) systems relying on multiple cooperative and non-cooperative tracking technologies to identify the risk of collision of unmanned aircraft systems (UAS) with other flight vehicles. In particular, fault tree analysis (FTA) is utilized to measure the overall system unavailability for each basic component failure. Considering the inter-dependencies of navigation and surveillance systems, the common cause failure (CCF)-beta model is applied to calculate the system risk associated with common failures. Additionally, an importance analysis is conducted to quantify the safety measures and identify the most significant component failures. Results indicate that the failure in traffic detection by cooperative surveillance systems contribute more to the overall DAA system functionality and that the probability of failure for ownship locatability in cooperative surveillance is greater than its traffic detection function. Although all the sensors individually yield 99.9% operational availability, the implementation of adequate multi-sensor DAA system relying on both cooperative and non-cooperative technologies is shown to be necessary to achieve the desired levels of safety in all possible encounters. These results strongly support the adoption of a unified analytical framework for cooperative/non-cooperative UAS DAA and elicits an evolution of the current certification framework to properly account for artificial intelligence and machine-learning based systems.

show abstract

“…Many reviews have been completed on relevant issues in the past two years [5][6][7][8]. Popular collision avoidance algorithms include geometric method [9][10][11], sampling-based method [12][13][14], numerical optimization [15][16][17][18][19], and artificial potential field (APF) [20][21][22][23][24][25][26][27][28]. The geometric method considers the geometric representation of the collision scene in the search for collision avoidance maneuvers.…”

Section: Introductionmentioning

confidence: 99%

Automatic Separation Management Between Multiple Unmanned Aircraft Vehicles in Uncertain Dynamic Airspace Based on Trajectory Prediction

Du¹,

Wang²,

Zhang³

et al. 2019

RIA

View full text Add to dashboard Cite

In an uncertain dynamic airspace, the future trajectories of noncooperative aircrafts (obstacles) are very uncertain. Multiple unmanned aircraft vehicles (UAVs) must avoid colliding into noncooperative obstacles and keep a safe distance between each other. This poses a huge challenge to the unmanned aircraft system traffic management (UTM). To cope with the challenge, this paper puts forward an automatic separation management method for a formation of multiple UAVs based on trajectory prediction in 3D dynamic airspace. Firstly, the uncertain trajectories of each obstacle in a time horizon were predicted based on the reachable set, producing an ellipsoidal reachable region. Next, a safe and efficient double-layer separation management strategy was proposed based on the improved artificial potential field (APF) method, considering the safe distance between cooperative UAVs and that between cooperative and noncooperative UAVs. The distance-based traditional APF method was adopted to manage the conflicts according to the reachable region of each obstacle, maximizing the safety between the UAV and the obstacle. The APF method based on relative motion state was employed to manage the conflicts, and adaptively adjust the dynamic safe separation between the UAVs. Experimental results prove that our method can effectively prevent the conflicts between a UAV and other UAVs in the formation or noncooperative obstacle. Besides, the collision-free trajectory generated by our method is smooth and stable, and close to the planned trajectory. The proposed method provides a solid guarantee to UAV flight safety, while minimizing the impacts on nearby aircrafts. The research findings shed new light on the UTM of highly uncertain low-altitude airspaces.

show abstract

Collision Avoidance Strategies for Unmanned Aerial Vehicles in Formation Flight

Cited by 130 publications

References 54 publications

Automated Enemy Avoidance of Unmanned Aerial Vehicles Based on Reinforcement Learning

Automated Enemy Avoidance of Unmanned Aerial Vehicles Based on Reinforcement Learning

Probabilistic Safety Assessment for UAS Separation Assurance and Collision Avoidance Systems

Automatic Separation Management Between Multiple Unmanned Aircraft Vehicles in Uncertain Dynamic Airspace Based on Trajectory Prediction

Contact Info

Product

Resources

About