This study presents a comprehensive survey of the current state-of-the-art techniques in virtual scene generation, particularly within the context of autonomous driving. The integration of deep learning methods such as generative adversarial networks (GANs) and convolutional LSTM (ConvLSTM) is explored in detail. Additionally, the effectiveness and applicability of these techniques in simulating real-world traffic scenarios are analyzed. Our article aims to bridge the gap between theoretical models and practical applications, providing an in-depth understanding of how deep learning and virtual scene generation converge to enhance the efficacy of autonomous driving systems

Ayyildiz, Dilara Vefa, Alnaser, Ala Jamil, Taj, Shahram, Zakaria, Mahta, Jaimes, Luis Gabriel

Heavy-Duty OBD IUMPR Data Collection Tool Process

J3162_202411 (Current)

11/21/2024

This SAE Information Report describes the collection of IUMPR data required by the heavy-duty onboard diagnostic regulation 13 CCR § 1971.1 (l)(2.3.3), using SAE J1939-defined messages incorporated in a suite of software functions

Truck and Bus Control and Communications Network Committee

Performance Requirements for Determining Tow-Vehicle Gross Combination Weight Rating and Trailer Weight Rating

J2807_202411 (Current)

11/21/2024

This document establishes minimum performance criteria at GCWR and calculation methodology to determine tow-vehicle TWR for passenger cars, multipurpose passenger vehicles, and trucks. This includes all vehicles up to 14000 pounds GVWR

Tow Vehicle Trailer Rating Committee

YOU’VE IMPLEMENTED ENCRYPTION, NOW WHAT

2024-01-3563

11/15/2024

Kline, Jonathan

RAPID HIGH-DIMENSIONAL SEMANTIC SEGMENTATION WITH ECHO STATE NETWORKS

2024-01-3939

11/15/2024

ABSTRACT Recurrent Neural Networks have largely been explored for low-dimensional time-series tasks due to their fading memory properties, which is not needed for feed-forward methods like the Convolutional Neural Network. However, benefits of using a recurrent-based neural network (i.e. reservoir computing) for time-independent inputs includes faster training times, lower training requirements, and reduced computational burdens, along with competitive performances to standard machine learning methods. This is especially important for high-dimensional signals like complex images. In this report, a modified Echo State Network (ESN) is introduced and evaluated for its ability to perform semantic segmentation. The parallel ESN containing 16 parallel reservoirs has an image processing time of 2 seconds with an 88% classification rate of 3 classes, with no prior feature extraction or normalization, and a training time of under 2 minutes. Citation: S. Gardner, M. R. Haider, J. Smereka, P

Gardner, S., Haider, M.R., Smereka, J., Jayakumar, P., Kulkarni, K., Gorsich, D., Moradi, L., Vantsevich, V.

LIDAR SEMANTIC SEGMENTATION WITH A MULTI-RESERVOIR ECHO STATE NETWORK FOR OFF-ROAD TERRAIN PERCEPTION

2024-01-3999

11/15/2024

ABSTRACT Autonomous vehicle perception has been widely explored using camera images but is limited with respect to LiDAR point cloud processing. Furthermore, focus is primarily on well-regulated environments, obviating a need for an algorithm that can contextualize dynamic and complex conditions through 3D point cloud representation. In this report, an Echo State Network for LiDAR signal processing is introduced and evaluated for its ability to perform semantic segmentation on unregulated terrains, using the RELLIS-3D open-source dataset. The L-ESN contains 16 parallel reservoirs with point cloud processing time of 1.9 seconds and 83.1% classification rate of 4 classes defining terrain trafficability, with no prior feature extraction or normalization, and a training time of 31 minutes. A 2D cost map is generated from the segmented point cloud for integration as a perception node plug-in to system-level navigation architectures. Citation: S. Gardner, M. R. Haider, P. Fiorini, S. Misko

Gardner, S., Haider, M. R., Fiorini, P., Misko, S., Smereka, J., Jayakumar, P., Gorsich, D., Moradi, L., Vantsevich, V.

THE UK’S APPROACH TO OPEN SYSTEMS ARCHITECTURE FOR THE LAND DOMAIN

2024-01-3441

11/15/2024

ABSTRACT This paper proposes that within the Land domain, there is not only a need to define an approach to open architectures, but also to mandate their use, in order to provide an agile framework for our fighting forces going forward. The paper sets out to explain such an approach; that taken by UK MOD and industry to produce the Generic Vehicle Architecture (GVA) defense standard. It will discuss how the GVA standard was formed, how it is currently being used and how it contributes to the wider MOD initiative for Open Systems Architecture for the Land domain. Finally the paper considers how the UK GVA relates to the US Victory standard and how interoperability may be achieved

White, Antony

An Open, ROS2, AGVRA-Based Autonomy Software Architecture for Military Robotic and Autonomous Systems

2024-01-3881

11/15/2024

ABSTRACT This paper reports on a prototype project to develop and mature a common, open, comprehensive, collaboratively developed, Government-owned, autonomy software architecture for ground robotic and autonomous systems (RAS). The prototype architecture, codenamed “SCION” promises the flexibility needed by the both the Government and industry research, development, testing, and engineering (RDT&E) communities to leverage reusable software and more rapidly innovate new capabilities; while ensuring the discipline and enabling the modularity required to develop RDT&E software structured to meet the software safety, cybersecurity, upgradeability, and other needs of RAS programs of record. Accordingly, program offices can adopt an acquisition strategy that requires compatibility with the de facto, military RAS standard, SCION architecture, while providing OEMs with baseline, SCION-compatible, GFE software (e.g. a future version of RTK). Such a strategy encourages optimal innovation

Thomasmeyer, William, John, Jon St., Martin, Dave, Mattes, Rich

BLOCKCHAIN INTEGRITY CHECKS FORSOFTWARE (BICS

2024-01-3983

11/15/2024

ABSTRACT Currently there is no method to ensure that the software loaded on a vehicle has been compromised at the software level. Common practice is to use physical port security to secure all network and data bus connection points with physical devices requiring tool, keys, or damage to tamper evident devices to prevent, inhibit, or discourage unauthorized connection; turn off access to the ports in the BIOS and password protect the BIOS. As well as give non-admin access to user accounts and password protect the operating systems. All these countermeasures help to prevent access but there is no way to tell if the software was compromised if not detected by these methods. Blockchain technology ensures that the software has not been compromised by comparing a hash generated at start up and comparing it to the distributed ledger. This technology helps to bring Warfighter technology into the future

Fortney, George G

Items per page:

1 – 50 of 5670