The fusion of LiDAR and camera sensors has demonstrated significant effectiveness in achieving accurate detection for short-range tasks in autonomous driving. However, this fusion approach could face challenges when dealing with long-range detection scenarios due to disparity between sparsity of LiDAR and high-resolution camera data. Moreover, sensor corruption introduces complexities that affect the ability to maintain robustness, despite the growing adoption of sensor fusion in this domain. We present SaViD, a novel framework comprised of a three-stage fusion alignment mechanism designed to address long-range detection challenges in the presence of natural corruption. The SaViD framework consists of three key elements: the Global Memory Attention Network (GMAN), which enhances the extraction of image features through offering a deeper understanding of global patterns; the Attentional Sparse Memory Network (ASMN), which enhances the inte-gration of LiDAR and image features; and the KNNnectivity Graph Fusion (KGF), which enables the entire fusion of spatial information. SaViD achieves superior performance on the long-range detection Argoverse-2 (AV2) dataset with a performance improvement of 9.87% in AP value and an improvement of 2.39% in mAPH for L2 difficulties on the Waymo Open dataset (WOD). Comprehensive experiments are carried out to showcase its robustness against 14 natural sensor corruptions. SaViD exhibits a robust performance improvement of 31.43% for AV2 and 16.13% for WOD in RCE value compared to other existing fusion-based methods while considering all the corruptions for both datasets. Our code is available at SaVil).

Combining LiDAR and camera data has shown potential in enhancing short-distance object detection in autonomous driving systems. Yet, the fusion encounters difficulties with extended distance detection due to the contrast between LiDAR’s sparse data and the dense resolution of cameras. Besides, discrepancies in the two data representations further complicate fusion methods. We introduce AYDIV, a novel framework integrating a tri-phase alignment process specifically designed to enhance long-distance detection even amidst data discrepancies. AYDIV consists of the Global Contextual Fusion Alignment Transformer (GCFAT), which improves the extraction of camera features and provides a deeper understanding of large-scale patterns; the Sparse Fused Feature Attention (SFFA), which fine-tunes the fusion of LiDAR and camera details; and the Volumetric Grid Attention (VGA) for a comprehensive spatial data fusion. AYDIV’s performance on the Waymo Open Dataset (WOD) with an improvement of 1.24% in mAPH value(L2 difficulty) and the Argoverse2 Dataset with a performance improvement of 7.40% in AP value demonstrates its efficacy in comparison to other existing fusion-based methods. Our code is publicly available at https://github.com/sanjay-810/AYDIV2

The continuous increase in air traffic compels major airports to optimize their resources and enhance their Terminal Maneuvering Airspace (TMA) operations. The primary constraint to increasing airport capacity is the required separation minima between pairs of aircraft arriving through the same approach routes. RECAT-EU is a revised global wake separation minima scheme released by EUROCONTROL in 2018. This suggested a reduction in separation minima between certain aircraft pairs while maintaining safety levels. However, it increases in separation minima scheme complexity by doubling the number of non-minimum radar separation (MRS) values. In this work, a Meta-Heuristic based optimization model is proposed to sequence arrival flights based on the RECAT-EU separation scheme, which can provide an optimized vectoring to ensure flights absorb their assigned delays before reaching the final approach fix. Findings show that the proposed model is able to generate an optimized landing sequence for 50 arrival flights in a computation time of 16 seconds. It also suggests that the proposed algorithm's computational time increases linearly with increasing the number of flights. Furthermore, trajectory vectoring results demonstrate that 85% of the assigned delays could be absorbed by flying the proposed vectored trajectories.

This paper presents a distributed reinforcement learning solution to the problem of dynamic channel allocation for cellular telecommunication networks in the presence of mobile call handoffs. The performance of various dynamic channel allocation schemes are compared via extensive computer simulations, and it is shown that a reduced-state SARSA reinforcement learning algorithm can achieve superior new call and handoff blocking probabilities. A new distributed reduced state SARSA algorithm is also developed which uses only local environment information readily available to the learning agent. By way of computer simulations, the distributed SARSA algorithm is shown to be capable of producing call blocking probabilities that are comparable to those obtained by the centralised learning agent.

This paper presents a reduced-state reinforcement learning solution to the dynamic channel allocation problem in cellular telecommunication networks featuring mobile traffic and call handoffs. We examine the performance of table-based function representation used in conjunction with the on-policy reinforcement learning algorithm SARSA and show that the policy obtained using a reduced-state table-based technique provides an online dynamic channel allocation solution with superior performance in terms of new call and handoff blocking probability as well as significantly reduced memory requirements. The superior performance of the proposed state-reduced technique is illustrated in simulation examples.