The unmanned aerial vehicle (UAV)-enabled communication technology is regarded as an efficient and effective solution for some special application scenarios where existing terrestrial infrastructures are overloaded to provide reliable services. To maximize the utility of the UAV-enabled system while meeting the QoS and energy constraints, the UAV needs to plan its trajectory considering the dynamic characteristics of scenarios, which is formulated as the Markov Decision Process (MDP). To solve the above problem, a deep reinforcement learning (DRL)-based scheme is proposed here, which predicts the trend of the dynamic scenarios to provide a long-term view for the UAV trajectory planning. Simulation results validate that our proposed scheme converges more quickly and achieves the better performance in dynamic scenarios.
相關內容
Unmanned Aerial Vehicles(UAVs) are attaining more and more maneuverability and sensory ability as a promising teleoperation platform for intelligent interaction with the environments. This work presents a novel 5-degree-of-freedom (DoF) unmanned aerial vehicle (UAV) cyber-physical system for aerial manipulation. This UAV's body is capable of exerting powerful propulsion force in the longitudinal direction, decoupling the translational dynamics and the rotational dynamics on the longitudinal plane. A high-level impedance control law is proposed to drive the vehicle for trajectory tracking and interaction with the environments. In addition, a vision-based real-time target identification and tracking method integrating a YOLO v3 real-time object detector with feature tracking, and morphological operations is proposed to be implemented onboard the vehicle with support of model compression techniques to eliminate latency caused by video wireless transmission and heavy computation burden on traditional teleoperation platforms.
Unmanned aerial vehicles (UAVs) can be applied in many Internet of Things (IoT) systems, e.g., smart farms, as a data collection platform. However, the UAV-IoT wireless channels may be occasionally blocked by trees or high-rise buildings. An intelligent reflecting surface (IRS) can be applied to improve the wireless channel quality by smartly reflecting the signal via a large number of low-cost passive reflective elements. This article aims to minimize the energy consumption of the system by jointly optimizing the deployment and trajectory of the UAV. The problem is formulated as a mixed-integer-and-nonlinear programming (MINLP), which is challenging to address by the traditional solution, because the solution may easily fall into the local optimal. To address this issue, we propose a joint optimization framework of deployment and trajectory (JOLT), where an adaptive whale optimization algorithm (AWOA) is applied to optimize the deployment of the UAV, and an elastic ring self-organizing map (ERSOM) is introduced to optimize the trajectory of the UAV. Specifically, in AWOA, a variable-length population strategy is applied to find the optimal number of stop points, and a nonlinear parameter a and a partial mutation rule are introduced to balance the exploration and exploitation. In ERSOM, a competitive neural network is also introduced to learn the trajectory of the UAV by competitive learning, and a ring structure is presented to avoid the trajectory intersection. Extensive experiments are carried out to show the effectiveness of the proposed JOLT framework.
Using Artificial Neural Networks (ANN) for nonlinear system identification has proven to be a promising approach, but despite of all recent research efforts, many practical and theoretical problems still remain open. Specifically, noise handling and models, issues of consistency and reliable estimation under minimisation of the prediction error are the most severe problems. The latter comes with numerous practical challenges such as explosion of the computational cost in terms of the number of data samples and the occurrence of instabilities during optimization. In this paper, we aim to overcome these issues by proposing a method which uses a truncated prediction loss and a subspace encoder for state estimation. The truncated prediction loss is computed by selecting multiple truncated subsections from the time series and computing the average prediction loss. To obtain a computationally efficient estimation method that minimizes the truncated prediction loss, a subspace encoder represented by an artificial neural network is introduced. This encoder aims to approximate the state reconstructability map of the estimated model to provide an initial state for each truncated subsection given past inputs and outputs. By theoretical analysis, we show that, under mild conditions, the proposed method is locally consistent, increases optimization stability, and achieves increased data efficiency by allowing for overlap between the subsections. Lastly, we provide practical insights and user guidelines employing a numerical example and state-of-the-art benchmark results.
The Internet of underwater things (IoUT) is envisioned to be an essential part of maritime activities. Given the IoUT devices' wide-area distribution and constrained transmit power, autonomous underwater vehicles (AUVs) have been widely adopted for collecting and forwarding the data sensed by IoUT devices to the surface-stations. In order to accommodate the diverse requirements of IoUT applications, it is imperative to conceive a multi-tier underwater computing (MTUC) framework by carefully harnessing both the computing and the communications as well as the storage resources of both the surface-station and of the AUVs as well as of the IoUT devices. Furthermore, to meet the stringent energy constraints of the IoUT devices and to reduce the operating cost of the MTUC framework, a joint environment-aware AUV trajectory design and resource management problem is formulated, which is a high-dimensional NP-hard problem. To tackle this challenge, we first transform the problem into a Markov decision process (MDP) and solve it with the aid of the asynchronous advantage actor-critic (A3C) algorithm. Our simulation results demonstrate the superiority of our scheme.
Interactive traffic simulation is crucial to autonomous driving systems by enabling testing for planners in a more scalable and safe way compared to real-world road testing. Existing approaches learn an agent model from large-scale driving data to simulate realistic traffic scenarios, yet it remains an open question to produce consistent and diverse multi-agent interactive behaviors in crowded scenes. In this work, we present InterSim, an interactive traffic simulator for testing autonomous driving planners. Given a test plan trajectory from the ego agent, InterSim reasons about the interaction relations between the agents in the scene and generates realistic trajectories for each environment agent that are consistent with the relations. We train and validate our model on a large-scale interactive driving dataset. Experiment results show that InterSim achieves better simulation realism and reactivity in two simulation tasks compared to a state-of-the-art learning-based traffic simulator.
In this study we consider unconditionally non-oscillatory, high order implicit time marching based on time-limiters. The first aspect of our work is to propose the high resolution Limited-DIRK3 (L-DIRK3) scheme for conservation laws and convection-diffusion equations in the method-of-lines framework. The scheme can be used in conjunction with an arbitrary high order spatial discretization scheme such as 5th order WENO scheme. It can be shown that the strongly S-stable DIRK3 scheme is not SSP and may introduce strong oscillations under large time step. To overcome the oscillatory nature of DIRK3, the key idea of L-DIRK3 scheme is to apply local time-limiters (K.Duraisamy, J.D.Baeder, J-G Liu), with which the order of accuracy in time is locally dropped to first order in the regions where the evolution of solution is not smooth. In this way, the monotonicity condition is locally satisfied, while a high order of accuracy is still maintained in most of the solution domain. For convenience of applications to systems of equations, we propose a new and simple construction of time-limiters which allows flexible choice of reference quantity with minimal computation cost. Another key aspect of our work is to extend the application of time-limiter schemes to multidimensional problems and convection-diffusion equations. Numerical experiments for scalar/systems of equations in one- and two-dimensions confirm the high resolution and the improved stability of L-DIRK3 under large time steps. Moreover, the results indicate the potential of time-limiter schemes to serve as a generic and convenient methodology to improve the stability of arbitrary DIRK methods.
In mixed-initiative co-creation tasks, where a human and a machine jointly create items, it is valuable for the generative system to provide multiple relevant suggestions to the designer. Quality-diversity algorithms have been commonly used for this, as they can provide diverse suggestions that are representative of salient areas of the solution space, showcasing solutions with both high fitness and different properties that the designer might be interested in. Since these suggestions are what drives the search process, it is important that they provide the right inspiration for the designer, as well as not stray too far away from the search trajectory, i.e., they should be aligned with what the designer is looking for. Additionally, in most cases, many interactions with the system are required before the designer is content with a solution. In this work, we tackle both of these problems with an interactive constrained MAP-Elites system by crafting emitters that are able to learn the preferences of the designer and use them in automated hidden steps. By learning such preferences, we remain aligned with the designer's intentions, and by applying automatic steps, we generate more solutions per system interaction, giving a larger number of choices to the designer and speeding up the search process. We propose a general framework for preference-learning emitters and test it on a procedural content generation task in the video game Space Engineers. In an internal study, we show that preference-learning emitters allow users to more quickly find relevant solutions.
Mobile edge devices see increased demands in deep neural networks (DNNs) inference while suffering from stringent constraints in computing resources. Split computing (SC) emerges as a popular approach to the issue by executing only initial layers on devices and offloading the remaining to the cloud. Prior works usually assume that SC offers privacy benefits as only intermediate features, instead of private data, are shared from devices to the cloud. In this work, we debunk this SC-induced privacy protection by (i) presenting a novel data-free model inversion method and (ii) demonstrating sample inversion where private data from devices can still be leaked with high fidelity from the shared feature even after tens of neural network layers. We propose Divide-and-Conquer Inversion (DCI) which partitions the given deep network into multiple shallow blocks and inverts each block with an inversion method. Additionally, cycle-consistency technique is introduced by re-directing the inverted results back to the model under attack in order to better supervise the training of the inversion modules. In contrast to prior art based on generative priors and computation-intensive optimization in deriving inverted samples, DCI removes the need for real device data and generative priors, and completes inversion with a single quick forward pass over inversion modules. For the first time, we scale data-free and sample-specific inversion to deep architectures and large datasets for both discriminative and generative networks. We perform model inversion attack to ResNet and RepVGG models on ImageNet and SNGAN on CelebA and recover the original input from intermediate features more than 40 layers deep into the network.
Human-centric perception plays a vital role in vision and graphics. But their data annotations are prohibitively expensive. Therefore, it is desirable to have a versatile pre-train model that serves as a foundation for data-efficient downstream tasks transfer. To this end, we propose the Human-Centric Multi-Modal Contrastive Learning framework HCMoCo that leverages the multi-modal nature of human data (e.g. RGB, depth, 2D keypoints) for effective representation learning. The objective comes with two main challenges: dense pre-train for multi-modality data, efficient usage of sparse human priors. To tackle the challenges, we design the novel Dense Intra-sample Contrastive Learning and Sparse Structure-aware Contrastive Learning targets by hierarchically learning a modal-invariant latent space featured with continuous and ordinal feature distribution and structure-aware semantic consistency. HCMoCo provides pre-train for different modalities by combining heterogeneous datasets, which allows efficient usage of existing task-specific human data. Extensive experiments on four downstream tasks of different modalities demonstrate the effectiveness of HCMoCo, especially under data-efficient settings (7.16% and 12% improvement on DensePose Estimation and Human Parsing). Moreover, we demonstrate the versatility of HCMoCo by exploring cross-modality supervision and missing-modality inference, validating its strong ability in cross-modal association and reasoning.
With the rapid increase of large-scale, real-world datasets, it becomes critical to address the problem of long-tailed data distribution (i.e., a few classes account for most of the data, while most classes are under-represented). Existing solutions typically adopt class re-balancing strategies such as re-sampling and re-weighting based on the number of observations for each class. In this work, we argue that as the number of samples increases, the additional benefit of a newly added data point will diminish. We introduce a novel theoretical framework to measure data overlap by associating with each sample a small neighboring region rather than a single point. The effective number of samples is defined as the volume of samples and can be calculated by a simple formula $(1-\beta^{n})/(1-\beta)$, where $n$ is the number of samples and $\beta \in [0,1)$ is a hyperparameter. We design a re-weighting scheme that uses the effective number of samples for each class to re-balance the loss, thereby yielding a class-balanced loss. Comprehensive experiments are conducted on artificially induced long-tailed CIFAR datasets and large-scale datasets including ImageNet and iNaturalist. Our results show that when trained with the proposed class-balanced loss, the network is able to achieve significant performance gains on long-tailed datasets.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191