Intelligent Dynamic Trajectory Planning of UAVs: Addressing Unknown Environments and Intermittent Target Loss

Drones, Journal Year: 2024, Volume and Issue: 8(12), P. 721 - 721

Published: Nov. 29, 2024

Precise trajectory planning is crucial in terms of enabling unmanned aerial vehicles (UAVs) to execute interference avoidance and target capture actions unknown environments when facing intermittent loss. To address the problem UAVs such conditions, this paper proposes a UAV system that includes predictor planner. Specifically, employs bidirectional Temporal Convolutional Network (TCN) Gated Recurrent Unit (GRU) network algorithm with an adaptive attention mechanism (BITCN-BIGRU-AAM) train model incorporates historical motion features intention inferred by Dynamic Bayesian (DBN). The resulting predictions maneuvering are used as terminal inputs for An improved Radial Basis Function (RBF) utilized combination offline–online method real-time obstacle planning. Additionally, considering future practical applications, planner adopt parallel optimization correction structure ensure accuracy computational efficiency. Simulation results indicate proposed can accurately avoid dynamic precisely during tasks involving loss, while also meeting capacity requirements.

Language: Английский

---

Lump-type kink wave phenomena of the space-time fractional phi-four equation

AIMS Mathematics, Journal Year: 2024, Volume and Issue: 9(12), P. 34372 - 34386

Published: Jan. 1, 2024

<p>This article examines the space-time fractional phi-four (PF) model which is an improvement of Klein-Fock-Gordon (KFG) that relevant in quantum mechanics, transmission ultra-short pulses optical fibers, de Broglie wave duality, and compound particles spinless relativistic systems. New solitary solutions equation were found with help Riccati-Bernoulli sub-ODE method together Bäcklund transformation. The conformable derivative played role improving modeling complex dynamical obtained solutions, illustrated by giving 2D, 3D MATLAB plots contour illustrations, s eful nuclear, particle physics fluid mechanics. Thus, current work demonstrates applicability proposed approach handling order deviations contributes to investigation dynamic behavior PF model.</p>

Language: Английский

---

自动驾驶开源数据体系：现状与未来

Scientia Sinica Informationis, Journal Year: 2024, Volume and Issue: unknown

Published: Feb. 1, 2024

随着自动驾驶技术的不断成熟与应用,系统性梳理开源自动驾驶数据集有利于产业生态良性循环。现有自动驾驶数据集可大致分为两代,第一代数据集的传感模态复杂度相对较低、数据集规模相对较小,且大多局限于感知级任务,第一代数据集以发布于 2012 年的 KITTI 为代表。相比于第一代数据集,第二代数据集的特征为传感模态复杂度较高、数据集规模与多样性较丰富、所设置任务从感知扩展到预测、规控上,第二代数据集以 2019 年前后提出的 nuScenes、Waymo 为代表。本评述联合学术界、产业界同仁,首次系统性梳理了国内外七十余种开源自动驾驶数据集,对如何构建高质量数据集、数据在算法闭环体系中发挥的核心作用、如何利用生成式大模型规模化生产数据等进行了总结。此外,就未来第三代自动驾驶数据集应该具备的特质和数据规模,以及需解决的科学与技术问题,我们进行了详细分析与讨论。本文的归纳与展望,有利于希望本文能促进新一代自动驾驶数据集与生态体系的建设、推动关键领域自主原创与科技自强的发展。更多内容请参见https://github.com/OpenDriveLab/DriveAGI。

---

Characterization and High‐Frequency Applications of Barium Antimonide Thin Films for Voltage‐Controlled Negative Capacitance and 6G Technology

physica status solidi (a), Journal Year: 2024, Volume and Issue: 221(20)

Published: Aug. 1, 2024

Herein, films of barium antimonides (BaSb) are fabricated by a thermal deposition technique under vacuum pressure 10 −5 mbar. BaSb exhibit orthorhombic lattice. Over wide range scans, the mostly stoichiometric, displaying atomic contents 50.17 and 49.83 at% for Ba Sb, respectively. Morphological analyses these show growth large grains with sizes ranging from 1.64 to 4.14 μm. Dynamic electrical measurements on Ag/BaSb/Ag conducted in frequency domain 0.01–1.80 GHz. The display features voltage‐controlled negative capacitance source (VCNC) resonance–antiresonance peaks at critical Lorentz models spectral indicate that they correspond high‐frequency value 2.8 nF, density oscillators 4 × cm −3 , scattering time constant τ = 100 ns. Additionally, variety cutoff spectra, making them suitable applications. varies 2.6 GHz 1.0 THz as driving increases VCNC sources, behavior, high make thin promising thin‐film transistors 6 G technology

Language: Английский

---

Location Privacy Protection Method for Networked Vehicles Based on K-Means and Shamir

2022 IEEE 11th Data Driven Control and Learning Systems Conference (DDCLS), Journal Year: 2024, Volume and Issue: unknown, P. 1020 - 1025

Published: May 17, 2024

Language: Английский

---