The task allocation of unmanned aerial vehicle (UAV) swarm is one of the practical problems for UAV's application and serves as the premise for tackling swarm's complex missions. By establishing a heterogeneous swarm task allocation problem featuring three types of UAVs, constraints such as UAV types, flight time and task execution sequence are considered. An objective function considering flight distance and task time of various types of UAVs is designed. Inspired by the hierarchical interaction behavior of pigeons, the hierarchical structure strategy is proposed and combined with the basic pigeon-inspired optimization (PIO) to improve the population's exploration ability. Simultaneously, the finite tolerance strategy is implemented to prevent individuals from falling into local optima due to inefficient explorations. Accordingly, hierarchy tolerance PIO (HTPIO) algorithm is proposed. Through comparing with other three algorithms across benchmark functions and two examples of swarm task allocation problem, HTPIO obtains the best results on more than half of benchmark functions and all examples. It is proved that HTPIO can effectively deal with complex optimization problems without increasing computational consumption and ensures the population always maintains a strong optimization ability throughout the process.

Unmanned aerial vehicle (UAV) swarm target defense is a crucial and practical issue about group decision-making and cooperative control in multiagent systems. Hawk-pigeon game architecture is presented in this article to guide UAV swarm target defense. Firstly, a 6-degree-of-freedom UAV model is employed, and the control command converter is derived, which could be widely used in the transformation from the second-order integrator model to the 6-DOF UAV model. Moreover, the attack tactics and pursuit strategies of the defender UAVs are proposed inspired by the hawk's hunting mechanism, and the dynamic model of the attacker UAVs is formulated inspired by pigeon group homing behavior. Furthermore, the victory zone of the hawk-pigeon game is analyzed using the time of interception and explicitly exhibited in isochrones means. The proposed method is scalable and adaptive, adopting distributed decision-making to support large-scale UAV swarm engagement. Finally, comparative experiments in different scenarios demonstrate the effectiveness of the proposed method over state-of-the-art target-attacker-defender game methods on the win rate, the number of captures, and activity time.

In this article, a boundary control scheme is proposed to suppress 2-D vibration of Euler–Bernoulli beam with output constraints and input saturation. The original partial differential equations (PDEs) model is transformed to a new form containing virtual control. Then a boundary controller is designed via the backstepping method to suppress the coupled vibration. The hyperbolic tangent function and Nussbaum function are employed to deal with the input saturation. A barrier Lyapunov function with time adjusting function is introduced to suppress the structural vibration of Euler–Bernoulli beam with arbitrary initial conditions. The disturbance observer is designed to deal with the unknown boundary disturbance. Finally, the simulation results show the effectiveness of the proposed vibration controller.

Fashion recommendation systems are highly desired by customers to find visually-collocated fashion items, such as clothes, shoes, bags, etc. While existing methods demonstrate promising results, they remain lacking in flexibility and diversity, e.g. assuming a fixed number of items or favoring safe but boring recommendations. In this paper, we propose a novel fashion collocation framework,Neural Graph Filtering, that models a flexible set of fashion items via a graph neural network. Specifically, we consider the visual embeddings of each garment as a node in the graph, and describe the inter-garment relationship as the edge between nodes. By applying symmetric operations on the edge vectors, this framework allows varying numbers of inputs/outputs and is invariant to their ordering. We further include a style classifier augmented with focal loss to enable the collocation of significantly diverse styles, which are inherently imbalanced in the training set. To facilitate a comprehensive study on diverse fashion collocation, we reorganize Amazon Fashion dataset with carefully designed evaluation protocols. We evaluate the proposed approach on three popular benchmarks, the Polyvore dataset, the Polyvore-D dataset, and our reorganized Amazon Fashion dataset. Extensive experimental results show that our approach significantly outperforms the state-of-the-art methods with over 10% improvements on the standard AUC metric. More importantly, 82.5% of the users prefer our diverse-style recommendations over other alternatives in a real-world perception study.

Generating 3D point clouds is challenging yet highly desired. This work presents a novel autoregressive model, PointGrow, which can generate diverse and realistic point cloud samples from scratch or conditioned on semantic contexts. This model operates recurrently, with each point sampled according to a conditional distribution given its previously-generated points, allowing inter-point correlations to be well-exploited and 3D shape generative processes to be better interpreted. Since point cloud object shapes are typically encoded by long-range dependencies, we augment our model with dedicated self-attention modules to capture such relations. Extensive evaluations show that PointGrow achieves satisfying performance on both unconditional and conditional point cloud generation tasks, with respect to realism and diversity. Several important applications, such as unsupervised feature learning and shape arithmetic operations, are also demonstrated.

Augmented Reality (AR) is known to enhance user experience, however, it remains under-adopted in industry. We present an AR interaction system improving human-machine coordination in Internet of Things (IoT) and Industry 4.0 applications including manufacturing and assembly, maintenance and safety, and other highly-interactive functions. A driver of slow adoption is the computational complexity and inaccuracy in localization and rendering digital content. AR systems may render digital content close to the associated physical objects, but traditional object recognition and localization modules perform poorly when tracking texture-less objects and complex shapes, presenting a need for robust and efficient digital content rendering techniques. We propose a method of improving IoT-AR by integrating Deep Learning with AR to increase accuracy and robustness of the target object localization module, taking both color and depth images as input and outputting the target's pose parameters. Quantitative and qualitative experiments prove this system's efficacy and show potential for fusing these emerging technologies in real-world applications.

We present an Augmented Reality (AR) visualization and interaction tool for users to control Internet of Things (IoT) devices with hand gestures. Today, smart IoT devices are becoming increasingly ubiquitous with diverse forms and functions, yet most user controls over them are still limited to mobile devices and web interfaces. Recently, AR has been developed rapidly, and provided immersive solutions to enhance user experience of applications in many fields. Its capability to create immersive interactions allows AR to improve the way smart devices are controlled via more direct visual feedback. In this paper, we create a functional prototype of one such system, enabling seamless interactions with sound and lighting systems through the use of augmented hand-controlled interaction panels. To interpret users' intentions, we implement a standard 2D convolution neural network (CNN) for real-time hand gesture recognition and deploy it within our system. Our prototype is also equipped with a simple but effective object detector which can identify target devices within a proper range by analyzing geometric features. We evaluate the performance of our system qualitatively and quantitatively and demonstrate it on two smart devices.

Point clouds provide a flexible geometric representation suitable for countless applications in computer graphics; they also comprise the raw output of most 3D data acquisition devices. While hand-designed features on point clouds have long been proposed in graphics and vision, however, the recent overwhelming success of convolutional neural networks (CNNs) for image analysis suggests the value of adapting insight from CNN to the point cloud world. Point clouds inherently lack topological information, so designing a model to recover topology can enrich the representation power of point clouds. To this end, we propose a new neural network module dubbed EdgeConv suitable for CNN-based high-level tasks on point clouds, including classification and segmentation. EdgeConv acts on graphs dynamically computed in each layer of the network. It is differentiable and can be plugged into existing architectures. Compared to existing modules operating in extrinsic space or treating each point independently, EdgeConv has several appealing properties: It incorporates local neighborhood information; it can be stacked applied to learn global shape properties; and in multi-layer systems affinity in feature space captures semantic characteristics over potentially long distances in the original embedding. We show the performance of our model on standard benchmarks, including ModelNet40, ShapeNetPart, and S3DIS.