In this work, we propose a method for estimating depth for an image of a monocular camera in order to avoid a collision for the autonomous flight of a drone. The highest flight speed of a drone is generally approximate 22.2 m/s, and long-distant depth information is crucial for autonomous flights since if the long-distance information is not available, the drone flying at high speeds is prone to collisions. However, long-range, measurable depth cameras are too heavy to be equipped on a drone. This work applies Pix2Pix, which is a kind of Conditional Generative Adversarial Nets (CGAN). Pix2Pix generates depth images from a monocular camera. Additionally, this work applies optical flow to enhance the accuracy of depth estimation. In this work, we propose a highly accurate depth estimation method that effectively embeds an optical flow map into a monocular image. The models are trained with taking advantage of AirSim, which is one of the flight simulators. AirSim can take both monocular and depth images over a hundred meter in the virtual environment, and our model generates a depth image that provides the long-distance information than images captured by a common depth camera. We evaluate accuracy and error of our proposed method using test images in AirSim. In addition, the proposed method is utilized for flight simulation to evaluate the effectiveness to collision avoidance. As a result, our proposed method is higher accuracy and lower error than a state of work. Moreover, our proposed method is lower collision than a state of work.

Delivery drones have been attracting attention as one of the promising technologies to deliver packages. Several research studies on routing problems specifically for drone delivery scenarios have extended Vehicle Routing Problems (VRPs). Most existing VRPs are based on Traveling Salesman Problems (TSPs) for minimizing the overall distance. On the other hand, VRPs for drone delivery have been aware of energy consumption due to the consideration of battery capacity. Despite hovering motions with loading packages accounting for a large portion of energy consumption since delivery drones need to hover with several packages, little research has been conducted on drone routing problems that aim at the minimization of overall flight times. In addition, flight time is strongly influenced by windy conditions such as headwinds and tailwinds. In this paper, we propose a VRP for drone delivery in which flight time is dependent on the weight of packages in a windy environment, called Flight Speed-aware Vehicle Routing Problem with Load and Wind (FSVRPLW). In this paper, flight speed changes depending on the load and wind. Specifically, a heavier package slows down flight speeds and a lighter package speeds up flight speeds. In addition, a headwind slows down flight speeds and a tailwind speed up flight speeds. We mathematically derived the problem and developed a dynamic programming algorithm to solve the problem. In the experiments, we investigate how much impact both the weight of packages and the wind have on the flight time. The experimental results indicate that taking loads and wind into account is very effective in reducing flight times. Moreover, the results of comparing the effects of load and wind indicate that flight time largely depends on the weight of packages.

Vision-based fall-detection methods have been previously studied but many have limitations in terms of practicality. Due to differences in rooms, users do not set the camera or sensors at the same height. However, few studies have taken this into consideration. Moreover, some fall-detection methods are lacking in terms of practicality because only standing, sitting and falling are taken into account. Hence, this study constructs a data set consisting of various daily activities and fall events and studies the effect of camera/sensor height on fall-detection accuracy. Each activity in the data set is carried out by eight participants in eight directions and taken with the depth camera at five different heights. Many related studies heavily depended on human segmentation by using Kinect SDK but this is not reliable enough. To address this issue, this study proposes Enhanced Tracking and Denoising Alex-Net (ETDA-Net) to improve tracking and denoising performance and classify fall and non-fall events. Experimental results indicate that fall-detection accuracy is affected by camera height, against which ETDA-Net is robust, outperforming traditional deep learning based fall-detection methods.

This paper presents a redesigned YOLO-based model tailored for small-object detection in drone applications. To enhance its performance in detecting small and blurry targets, this study introduces the C3_CAA module to refine feature maps, integrates the CPA module and SI-IoU to improve detection accuracy, and incorporates channel and spatial attention mechanisms to further enhance target localization and identification performance.The experimental results indicated that the proposed method performs well on multiple datasets. The mAP value increases by 2% on the VISDRONE dataset, 1.6% on the UAVDT dataset, 0.9% on the CARPK dataset, and 1% on the UAVROD data set.

Detection of small or distant objects is a major challenge in 3-D object detection in autonomous driving either through RGB images or LiDAR point clouds. Despite the growing popularity of sensor fusion in this task, existing fusion methods have not adequately taken into account the challenges associated with 3-D small object detection, such as semantic misalignment of small objects, caused by occlusion and calibration errors. To address this issue, we propose dynamic point-pixel feature alignment network (DPPFA-Net) for multimodal 3-D small object detection by introducing memory-based point-pixel fusion (MPPF) modules, deformable point-pixel fusion (DPPF) modules, and semantic alignment evaluator (SAE) modules. More concretely, the proposed MPPF module automatically performs intramodal and cross-modal feature interactions. The intramodal interaction reduces sensitivity to noise points, while the explicit cross-modal feature interaction based on the memory bank facilitates easier network learning and enables a more comprehensive and discriminative feature representation. The DPPF module establishes interactions exclusively with key position pixels based on a sampling strategy. This design not only guarantees a low-computational complexity but also enables adaptive fusion functionality, especially beneficial for high-resolution images. The SAE module guarantees semantic alignment of the fused features, thereby enhancing the robustness and reliability of the fusion process. Furthermore, we construct a simulated multimodal noise data set, which enables quantitative analysis of the robustness of multimodal methods under varying degrees of multimodal noise. Extensive experiments on the KITTI benchmark and challenging multimodal noisy cases show that DPPFA-Net achieves a new state-of-the-art, highlighting its effectiveness in detecting small objects. Our proposed method is compared to the first place on the KITTI leaderboard and achieves better performance by 2.07%, 6.52%, 7.18%, and 6.22% of the average precision on the varying degrees of multimodal noise cases.

Existing models achieve high accuracy in small object detection but often overlook the rotational and dense characteristics of targets. To address this, we propose an improved YOLO11-based model specifically designed for detecting dense and rotated small objects in UAV scenarios. To better extract features from rotated targets, we design a C3k2–ARC module that enhances the model’s rotational detection capability. In addition, we introduce the CL-Concat feature fusion module, which combines traditional concatenation with channel and spatial attention, significantly improving the quality of multi-scale feature fusion. The experimental results demonstrate that the proposed method achieves notable performance improvements across multiple public benchmark data sets. Compared with the advanced YOLO11n model, our approach achieves gains of 1.6% on VISDRONE, 1.4% on UAVDT, 0.2% on CARPK, and 0.1% on UAVROD, further validating its effectiveness across diverse UAV detection scenarios.