Visual-based obstacle detection and tracking, and conflict detection for small UAS sense and avoid

Opromolla, Roberto; Fasano, Giancarmine

doi:10.1016/j.ast.2021.107167

Cited by 32 publications

(19 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Traditional sensors used for surveillance and tracking include ADS-B, onboard radar, and transponders [19]; however, autonomous aircraft require additional sensors both for redundancy and to replace the visual acquisition typically performed by the pilot. For this reason, vision-based traffic detection systems have been proposed, in which intruding aircraft are detected from images taken by a camera sensor mounted on the aircraft [5], [20]. In this section, we will apply our risk-driven design techniques to improve the safety of a vision-based detect and avoid (DAA) system.…”

Section: Vision-based Detect and Avoid Applicationmentioning

confidence: 99%

“…The design of reliable perception systems is a key challenge in the development of safety-critical autonomous systems [1], [2]. Modern perception systems are often required to predict state information from complex, high-dimensional inputs such as images or LiDAR data [3]- [5]. This information is then passed to a controller, which uses the state estimate to make safety-critical decisions.…”

Section: Introductionmentioning

confidence: 99%

“…This information is then passed to a controller, which uses the state estimate to make safety-critical decisions. For example, vision-based perception systems have been proposed for detect and avoid applications in aviation [5]. It is important that perception systems produce accurate estimates, and these systems are typically trained to minimize overall perceptual error using a regression loss on a set of training data.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Risk-Driven Design of Perception Systems

Corso¹,

Katz²,

Innes³

et al. 2022

Preprint

View full text Add to dashboard Cite

Modern autonomous systems rely on perception modules to process complex sensor measurements into state estimates. These estimates are then passed to a controller, which uses them to make safety-critical decisions. It is therefore important that we design perception systems to minimize errors that reduce the overall safety of the system. We develop a risk-driven approach to designing perception systems that accounts for the effect of perceptual errors on the performance of the fullyintegrated, closed-loop system. We formulate a risk function to quantify the effect of a given perceptual error on overall safety, and show how we can use it to design safer perception systems by including a risk-dependent term in the loss function and generating training data in risk-sensitive regions. We evaluate our techniques on a realistic vision-based aircraft detect and avoid application and show that risk-driven design reduces collision risk by 37 % over a baseline system. * equal contribution Preprint. Under review.

show abstract

Section: Vision-based Detect and Avoid Applicationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Risk-Driven Design of Perception Systems

Corso¹,

Katz²,

Innes³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…For instance, PP‐YOLO (YOLO based on Paddle)[8] achieves a balance between detection speed and precision on the basis of YOLOv3 [9], while YOLOv5 is an improvement of YOLOv4 to select different backbone networks in various missions based on their respective requirements for detection precision and timeliness. Recent research on UAV detection has been mainly applied to UAV collision avoidance and path planning [5, 6] and they mainly extracted UAV position information through DL‐based target detection algorithms; however, they did not consider how to pre‐process the video information to improve detection accuracy under poor field‐of‐view conditions.…”

Section: Introductionmentioning

confidence: 99%

“…For the safety of surface ships, UAVs must be quickly detected, protected against, and intercepted if encountered at sea. UAVs are mainly detected in such ways as radar [1], acoustic features [2], radio frequency [3,4], and video images [5,6]. The detection based on radar features long distance and good precision, but performs poorly in detecting and tracking the objects of low altitude, slow speed, and small size such as UAVs.…”

Section: Introductionmentioning

confidence: 99%

Unmanned aerial vehicles object detection based on image haze removal under sea fog conditions

Wang

Jiangxin

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) have gradually become a major air threat to ships because of small size, good maneuverability, and low cost. Vision-based UAV detection offers one of the main ways to identify and protect against UAVs. Unlike land environment, the weather is complicated at sea. The visibility of an object is undermined by such factors as sea fog and sunlight, which makes it difficult to detect UAVs at sea through visionbased object detection. For the purpose of object detection at sea, this paper proposes a UAV object detection method based on image haze removal. In the proposed method, an improved dark channel haze removal (DCHR) algorithm is utilized to remove haze for and restore video images. Additionally, co-ordinate attention (CoordAttention, CA) is introduced to the lightweight algorithms of You Only Look Once (YOLO) for the object detection in restored video images, so as to improve the precision and speed of detection and reduce the miss rate. Some video images are also taken for detection experiments to verify the feasibility and effectiveness of the proposed method.

show abstract

Developing an expansion‐based obstacle detection using panoptic segmentation

Pirasteh,

Varshosaz,

Badrloo

et al. 2024

Journal of Field Robotics

View full text Add to dashboard Cite

Safe Micro Aerial Vehicle (MAV) navigation requires detecting and avoiding obstacles. For safe MAV navigation, expansion‐based algorithms are effective for detecting obstacles. However, accurate and real‐time obstacle detection is a fundamental challenge. Some traditional methods focus on extracting geometric features from images and applying geometric constraints to identify potential obstacles. Others may leverage machine learning algorithms for object detection and classification, using features such as texture, shape, and context to distinguish obstacles from background clutter. The choice of approach depends on factors such as the specific requirements of the application, the complexity of the scene, and the available computational resources. Since obstacles, in reality, take the form of objects (e.g., persons, walls, pillars, trees, automobiles, and other structures), it is preferable to represent them according to human comprehension and as objects. Therefore, the objective of this study is to reflect on the previous research and address the issues mentioned above by extracting objects from the fisheye image using a panoptic deep‐learning network. The extracted object regions are, then, used to identify obstacles with a novel area‐based expansion rate we developed in a previous study. We compared the accuracy of obstacle detection in our proposed method to the existing method when moving forward and to the right; thus, we improved it between 10% and 18%, respectively. In addition, compared with the existing method, and due to replacing a single object with multiple regions, obstacle‐detection runtime for forward and right direction is 15.71 and 25.5 times faster, respectively, and the required match points have decreased by 49% and 55%.

show abstract

Visual-based obstacle detection and tracking, and conflict detection for small UAS sense and avoid

Cited by 32 publications

References 50 publications

Risk-Driven Design of Perception Systems

Risk-Driven Design of Perception Systems

Unmanned aerial vehicles object detection based on image haze removal under sea fog conditions

Developing an expansion‐based obstacle detection using panoptic segmentation

Contact Info

Product

Resources

About