Vision-Assisted Landing Method for Unmanned Powered Parachute Vehicle Based on Lightweight Neural Network

Zhang, Mengxuan; Hu, Wei; Ji, Shude; Song, Qi; Gong, Peng; Kong, Lingpei

doi:10.1109/access.2021.3112185

Cited by 5 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, our system is vision-agnostic and any detection method (e.g. [19], [20], [21]) can be used, depending on the specific application requirements. Following up the time-step approach explained earlier, we only target 8FPS instead of the 30FPS that the camera provides, further reducing the computational strain.…”

Section: A Target Detection and Parameter Estimationmentioning

confidence: 99%

Towards a Fully Autonomous UAV Controller for Moving Platform Detection and Landing

Piponidis

Aristodemou

Theocharides

2022

2022 35th International Conference on VLSI Design and 2022 21st International Conference on Embedded Systems (VLSID)

View full text Add to dashboard Cite

While Unmanned Aerial Vehicles (UAVs) are increasingly deployed in several missions, their inability of reliable and consistent autonomous landing poses a major setback for deploying such systems truly autonomously. In this paper we present an autonomous UAV landing system for landing on a moving platform. In contrast to existing attempts, the proposed system relies only on the camera sensor, and has been designed as lightweight as possible. The proposed system can be deployed on a low power platform as part of the drone payload, whilst being indifferent to any external communication or any other sensors. The system relies on a Neural Network (NN) based controller, for which a target and environment agnostic simulator was created, used in training and testing of the proposed system, via Reinforcement Learning (RL) and Proximal Policy Optimization (PPO) to optimally control and steer the drone towards landing on the target. Through real-world testing, the system was evaluated with an average deviation of 15cm from the center of the target, for 40 landing attempts.

show abstract

Section: A Target Detection and Parameter Estimationmentioning

confidence: 99%

Towards a Fully Autonomous UAV Controller for Moving Platform Detection and Landing

Piponidis

Aristodemou

Theocharides

2022

2022 35th International Conference on VLSI Design and 2022 21st International Conference on Embedded Systems (VLSID)

View full text Add to dashboard Cite

show abstract

“…At the same time, based on the kernel correlation filtering, the complementary filtering fusion detection algorithm was used to further reduce the detection error and improve computational efficiency. To achieve the accurate landing of UAVs in complex environments, a landing runway recognition model based on a deep learning algorithm was proposed by [ 27 ]. In this lightweight landing runway recognition model, the lightweight network Mobilenet-V3 was used to replace the underlying network structure CSPDarkNet-53 in the YOLOv4 algorithm, which meets the lightweight requirement for use on mobile devices.…”

Section: Related Workmentioning

confidence: 99%

Monocular-Vision-Based Precise Runway Detection Applied to State Estimation for Carrier-Based UAV Landing

Weng

Cao

et al. 2022

Sensors

View full text Add to dashboard Cite

Improving the level of autonomy during the landing phase helps promote the full-envelope autonomous flight capability of unmanned aerial vehicles (UAVs). Aiming at the identification of potential landing sites, an end-to-end state estimation method for the autonomous landing of carrier-based UAVs based on monocular vision is proposed in this paper, which allows them to discover landing sites in flight by using equipped optical sensors and avoid a crash or damage during normal and emergency landings. This scheme aims to solve two problems: the requirement of accuracy for runway detection and the requirement of precision for UAV state estimation. First, we design a robust runway detection framework on the basis of YOLOv5 (you only look once, ver. 5) with four modules: a data augmentation layer, a feature extraction layer, a feature aggregation layer and a target prediction layer. Then, the corner prediction method based on geometric features is introduced into the prediction model of the detection framework, which enables the landing field prediction to more precisely fit the runway appearance. In simulation experiments, we developed datasets applied to carrier-based UAV landing simulations based on monocular vision. In addition, our method was implemented with help of the PyTorch deep learning tool, which supports the dynamic and efficient construction of a detection network. Results showed that the proposed method achieved a higher precision and better performance on state estimation during carrier-based UAV landings.

show abstract

“…Therefore, scientifically planning a high-quality homing trajectory is the premise for the parafoil airdrop system to achieve reliable autonomous flight. It is also an essential guarantee for the final realization of an accurate and safe airdrop [6][7][8][9].…”

Section: Introductionmentioning

confidence: 99%

A Novel Trajectory Planning Method for Parafoil Airdrop System Based on Geometric Segmentation Strategy

Gao¹,

Jin²

2022

Preprint

View full text Add to dashboard Cite

Reasonable trajectory planning is the precondition for the parafoil airdrop system to achieve autonomous accurate homing and safe landing. In order to successfully realize the self-homing of the parafoil airdrop system, a new trajectory optimization design scheme is proposed in this paper. The scheme is based on the parafoil's unique flight and control characteristics and adopts a segmented homing design. Compared with the current common trajectory design method, its core feature is to avoid the problem that the straight-line flight distance before landing is limited by the radius of the height-reducing area and ensure landing accuracy and safety. Firstly, the different starting states of the parafoil airdrop system and the landing requirements were comprehensively considered, and the homing trajectory is reasonably segmented. Based on the requirements of energy control, stable flight, and landing accuracy, the optimal objective function of the trajectory was established, and the trajectory parameters, calculation methods, and constraints were given. Secondly, the cuckoo search algorithm was applied to optimize the objective function to obtain the final home trajectory. Finally, the trajectory planning under different airdrop conditions was simulated and verified. The results showed that the planned trajectories could reach the target point accurately and meet the flight direction requirements, proving the proposed scheme's correctness and feasibility.

show abstract

Vision-Assisted Landing Method for Unmanned Powered Parachute Vehicle Based on Lightweight Neural Network

Cited by 5 publications

References 27 publications

Towards a Fully Autonomous UAV Controller for Moving Platform Detection and Landing

Towards a Fully Autonomous UAV Controller for Moving Platform Detection and Landing

Monocular-Vision-Based Precise Runway Detection Applied to State Estimation for Carrier-Based UAV Landing

A Novel Trajectory Planning Method for Parafoil Airdrop System Based on Geometric Segmentation Strategy

Contact Info

Product

Resources

About