Browse Topic: Artificial intelligence (AI)

Items (1,811)
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 VefaAlnaser, Ala JamilTaj, ShahramZakaria, MahtaJaimes, Luis Gabriel
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.
ABSTRACT Future autonomous combat vehicles will need to travel off-road through poorly mapped environments. Three-dimensional topography may be known only to a limited extent (e.g. coarse height), but this will likely be noisy and of limited resolution. For ground vehicles, 3D topography will impact how far ahead the vehicle can “see”. Higher vantage points and clear views provide much more useful path planning data than lower vantage points and occluded views from trees and structures. The challenge is incorporating this knowledge into a path planning solution. When should the robot climb higher to get a better view or else continue moving along the shortest path predicted by current information? We investigated the use of Deep Q-Networks (DQN) to reason over this decision space, comparing performance to conventional methods. In the presence of significant sensor noise, the DQN was more successful in finding a path to the target than A* for all but one type of terrain. Citation: E
Martinson, EricPurman, BenDallas, Andy
ABSTRACT The effective and safe use of Rough Terrain Cargo Handlers is severely hampered by the operator’s view being obstructed. This results in the inability to see a) in front of the vehicle while driving, b) where to set a carried container, and c) where to maneuver the vehicles top handler in order to engage with cargo containers. We present an analysis of these difficulties along with specific solutions to address these challenges that go beyond the non-technical solution currently used, including the placement of sensors and the use of image analysis. These solutions address the use of perception to support autonomy, drive assist, active safety, and logistics
Beach, GlennHaanpaa, DouglassMoody, GaryMahal, PritpaulRowe, SteveSiebert, GaryBurkowski, JimCohen, Charles J.
ABSTRACT An increasing pace of technology advancements and recent heavy investment by potential adversaries has eroded the Army’s overmatch and spurred significant changes to the modernization enterprise. Commercial ground vehicle industry solutions are not directly applicable to Army acquisitions because of volume, usage and life cycle requirement differences. In order to meet increasingly aggressive schedule goals while ensuring high quality materiel, the Army acquisition and test and evaluation communities need to retain flexibility and continue to pursue novel analytic methods. Fully utilizing test and field data and incorporating advanced techniques, such as, big data analytics and machine learning can lead to smarter, more rapid acquisition and a better overall product for the Soldier. Logistics data collections during operationally relevant events that were originally intended for the development of condition based maintenance procedures in particular have been shown to provide
Heine, RichardFrounfelker, BradSalins, LaneWang, Chongying
ABSTRACT The armor research and development community needs a more cost-effective, science-based approach to accelerate development of new alloys (and alloys never intended for ballistic protection) for armor applications, especially lightweight armor applications. Currently, the development and deployment of new armor alloys is based on an expert-based, trial-and-error process, which is both time-consuming and costly. This work demonstrates a systematic research approach to accelerate optimization of the thermomechanical processing (TMP) pathway, yielding optimal microstructure and maximum ballistic performance. Proof-of-principle is being performed on titanium alloy, Ti-10V-2Fe-3Al, and utilizes the Hydrawedge® unit of the Gleeble 3800 System (a servo-hydraulic thermomechanical testing device) to quickly evaluate mechanical properties and simulate rolling schedules on small samples. Resulting mechanical property and microstructure data are utilized in an artificial intelligence (AI
Lillo, ThomasChu, HenryAnderson, JeffreyWalleser, JasonBurguess, Victor
ABSTRACT Lidar, Sonar, and Vision-based measurements are often used to preview terrain topology for unmanned ground vehicles. Environmental conditions such as wet or snow-covered roads, shadows, superficial ground coverings, and deceptive surface textures can lead to erroneous measurements. Tactile terrain prediction is both an alternative and a supplement to existing measurement systems. Tactile feedback from an array of low-cost sensors on the moving vehicle is used to generate low wave-number terrain profile predictions. This paper presents tactile terrain prediction results evaluated on four unique courses. Prediction error data are presented up to 25m in front of the vehicle. Results indicate 0.02-0.2m RMS error and 0.18-1.0m peak error at a 10m look-ahead distance. As expected, the prediction errors decrease exponentially as the look-ahead distance decreases. The relatively small prediction errors suggest that the proposed tactile terrain prediction method is a viable low-cost
Southward, Steve
ABSTRACT With recent advancements in the automotive world and the introductions of autonomous vehicles, automotive cybersecurity has become a main and primary issue for every automaker. In order to come up with measures to detect and protect against malicious attacks, intrusion detection systems (IDS) are commonly used. These systems identify attacks while comparing normal behavior with abnormalities. In this paper, we propose a novel, two-stage IDS based on deep-learning and rule-based systems. The objective of this IDS is to detect malicious attacks and ensure CAN security in real time. Deep Learning has already been used in CAN IDS and is already proven to be a successful algorithm when it comes to extensive datasets but comes with the cost of high computational requirements. The novelty of this paper is to use Deep Learning to achieve high predictability results while keeping low computational requirements by offsetting it with rule-based systems. In addition, we examine the
Zhang, LinxiKaja, NevrusShi, LyndonMa, Di
ABSTRACT In this paper we present an intelligent power controller for a vehicle power system that employs multiple power sources. In particular we focus on a vehicle power system architecture that is used in vehicles such as Mine Resistant Ambush Protected (MRAP) vehicle. These vehicles are designed to survive IED (Improvised Explosive Devices) attacks and ambushes. The power system has the following major components: a “clean” bus, a “dirty” bus, an engine, a hydraulic system and a switch between the clean and the dirty bus. We developed algorithms for intelligent energy management for this type of vehicle power system including DP (Dynamic Programming) optimization, DP online control and a machine learning technique that combines neural networks with DP to train an intelligent power controller. We present experiments conducted through modeling and simulation using a generic commercial software tool and a lab hardware setup
Chen, ZhihangMurphey, Yi L.Chen, ZhengMasrur, AbulMi, Chris
ABSTRACT Optical distortion measurements for transparent armor (TA) solutions are critical to ensure occupants can see what is happening outside a vehicle. Unfortunately, optically transparent materials often have poorer mechanical properties than their opaque counterparts which usually results in much thicker layups to provide the same level of protection. Current standards still call for the use of a double exposure method to manually compare the distortion of grid lines. This report presents provides a similar method of analysis with less user input using items typically available in many mechanics labs: machine vision cameras and digital image correlation software. Citation: J. M. Gorman, “An Easier Approach to Measuring Optical Distortion in Transparent Armor”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI, Aug. 11-13, 2020. The views presented are those of the author and do not necessarily represent the views of DoD or
Gorman, James M.
ABSTRACT The fundamental aspect of unmanned ground vehicle (UGV) navigation, especially over off-road environments, are representations of terrain describing geometry, types, and traversability. One of the typical representations of the environment is digital surface models (DSMs) which efficiently encode geometric information. In this research, we propose a collaborative approach for UGV navigation through unmanned aerial vehicle (UAV) mapping to create semantic DSMs, by leveraging the UAV wide field of view and nadir perspective for map surveying. Semantic segmentation models for terrain recognition are affected by sensing modality as well as dataset availability. We explored and developed semantic segmentation deep convolutional neural networks (CNN) models to construct semantic DSMs. We further conducted a thorough quantitative and qualitative analysis regarding image modalities (between RGB, RGB+DSM and RG+DSM) and dataset availability effects on the performance of segmentation
Brand, Howard J. J.Li, Bing
ABSTRACT This paper presents a method to mitigate high latency in the teleoperation of unmanned ground systems through display prediction and state estimation. Specifically, it presents a simulation environment which models both sides of the teleoperation system in the laboratory. The simulation includes a teleoperated vehicle model to represent the dynamics in high fidelity. The sensors and actuators are modeled as well as the communication channel. The latency mitigation approach is implemented in this simulation environment, which consists of a feed-forward vehicle model as a state estimator which drives a predictive display algorithm. These components work together to help the operator receive immediate feedback regarding his/her control actions. The paper contains a technical discussion of the design as well as specific implementation. It concludes with the presentation of some experimental data which demonstrate significant improvement over the unmitigated case
Brudnak, Mark J.
ABSTRACT An approach for a perception system for autonomous vehicle navigation is presented. The approach relies on low-cost electro-optical (EO) sensors for terrain classification, 3D environment modeling, and object/obstacle recognition. Stereo vision is used to generate real-time range maps which are populated into a hybrid probabilistic environment model. Textural and spectral cues are utilized for terrain classification and spatial contextual knowledge is proposed to augment object recognition performance
Flannigan, William C.Rigney, Michael P.Alley, Kevin J.
ABSTRACT Simulation is a critical step in the development of autonomous systems. This paper outlines the development and use of a dynamically linked library for the Mississippi State University Autonomous Vehicle Simulator (MAVS). The MAVS is a library of simulation tools designed to allow for real-time, high performance, ray traced simulation capabilities for off-road autonomous vehicles. It includes features such as automated off-road terrain generation, automatic data labeling for camera and LIDAR, and swappable vehicle dynamics models. Many machine learning tools today leverage Python for development. To use these tools and provide an easy to use interface, Python bindings were developed for the MAVS. The need for these bindings and their implementation is described. Citation: C. Hudson, C. Goodin, Z. Miller, W. Wheeler, D. Carruth, “Mississippi State University Autonomous Vehicle Simulation Library”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium
Hudson, Christopher R.Goodin, ChristopherMiller, ZachWheeler, WarrenCarruth, Daniel W.
ABSTRACT Model Based Systems Engineering (MBSE) has been a dominant methodology for defining and developing complex systems; however, it has not yet been paired with cutting-edge digital engineering transformation. MBSE is constrained to represent a whole system, but lacks other capabilities, such as dynamic simulation and optimization, as well as integration of hardware and software functions. This paper provides the key elements for developing a Smart MBSE (SMBSE) modeling approach that integrates Systems Engineering (SE) functionality with the full suite of other development tools utilized to create today’s complex products. SMBSE connects hardware and software with a set of customer needs, design requirements, program targets, simulations and optimization functionalities. The SMBSE modeling approach is still under development, with significant challenges for building bridges between conventional Systems Engineering methodology, with additional capabilities to reuse, automate
Ayala, AlejandroWeaver, JonathanFuentes, JeniferOchoa, Ruben
ABSTRACT The IGVC offers a design experience that is at the very cutting edge of engineering education, with a particular focus in developing engineering control/sensor integration experience for the college student participants. A main challenge area for teams is the proper processing of all the vehicle sensor feeds, optimal integration of the sensor feeds into a world map and the vehicle leveraging that world map to plot a safe course using robust control algorithms. This has been an ongoing challenge throughout the 26 year history of the competition and is a challenge shared with the growing autonomous vehicle industry. High consistency, reliability and redundancy of sensor feeds, accurate sensor fusion and fault-tolerant vehicle controls are critical, as even small misinterpretations can cause catastrophic results, as evidenced by the recent serious vehicle crashes experienced by self-driving companies including Tesla and Uber Optimal control techniques & sensor selection
Kosinski, AndrewIyengar, KiranTarakhovsky, JaneLane, JerryCheok, KaCTheisen, BernieOweis, Sami
ABSTRACT Recent advances in neuroscience, signal processing, machine learning, and related technologies have made it possible to reliably detect brain signatures specific to visual target recognition in real time. Utilizing these technologies together has shown an increase in the speed and accuracy of visual target identification over traditional visual scanning techniques. Images containing a target of interest elicit a unique neural signature in the brain (e.g. P300 event-related potential) when detected by the human observer. Computer vision exploits the P300-based signal to identify specific features in the target image that are different from other non-target images. Coupling the brain and computer in this way along with using rapid serial visual presentation (RSVP) of the images enables large image datasets to be accurately interrogated in a short amount of time. Together this technology allows for potential military applications ranging from image triaging for the image analyst
Ries, Anthony J.Lance, BrentSajda, Paul
ABSTRACT Motion planning algorithms for vehicles in an offroad environment have to contend with the significant vertical motion induced by the uneven terrain. Besides the obvious problems related to driver comfort, for autonomous vehicles, such “bumpy” vertical motion can induce significant mechanical noise in the real time data acquired from onboard sensors such as cameras to the point that perception becomes especially challenging. This paper advances a framework to address the problem of vertical motion in offroad autonomous motion control for vehicular systems. This framework is first developed to demonstrate the stabilization of the sprung mass in a modified quarter-car tracking a desired velocity while traversing a terrain with changing height. Even for an idealized model such as the quarter-car the dynamics turn out to be nonlinear and a model-based controller is not obvious. We therefore formulate this control problem as a Markov decision process and solve it using deep
Salvi, AmeyaBuzhardt, JakeTallapragada, PhanindraKrovi, VenkatBrudnak, MarkSmereka, Jonathon M.
ABSTRACT To realize the full potential of simulation-based evaluation and validation of autonomous ground vehicle systems, the next generation of modeling and simulation (M&S) solutions must provide real-time closed-loop environments that feature the latest physics-based modeling approaches and simulation solvers. Real-time capabilities enable seamless integration of human-in/on-the-loop training and hardware-in-the-loop evaluation and validation studies. Using an open modular architecture to close the loop between the physics-based solvers and autonomy stack components allows for full simulation of unmanned ground vehicles (UGVs) for comprehensive development, training, and testing of artificial intelligence vehicle-based agents and their human team members. This paper presents an introduction to a Proof of Concept for such a UGV M&S solution for severe terrain environments with a discussion of simulation results and future research directions. This conceptual approach features: 1
Misko, SamuelFree, ArnoldSivashankar, ShivaKluge, TorstenVantsevich, VladimirHirshkorn, MartinMorales, AndresBrascome, James MichaelRose, ShaylaBowen, NicZhang, SiyanGhasemi, MasoodGardner, StevenFiorini, PierreMaddela, MadhurimaJayakumar, ParamsothyGorsich, DavidManning, ChrisThurau, MatthiasRueddenklau, NicoZachariah, GibinDennis, EvaCostello, Ian
ABSTRACT This paper describes the use of neural networks to enhance simulations for subsequent training of anomaly-detection systems. Simulations can provide edge conditions for anomaly detection which may be sparse or non-existent in real-world data. Simulations suffer, however, by producing data that is “too clean” resulting in anomaly detection systems that cannot transition from simulated data to actual conditions. Our approach enhances simulations using neural networks trained on real-world data to create outputs that are more realistic and variable than traditional simulations. Citation: P.Feldman, “Training robust anomaly detection using ML-Enhanced simulations”, In Proceedings of the Ground Vehicle Systems Engineering and Technology Symposium (GVSETS), NDIA, Novi, MI, Aug. 11-13, 2020
Feldman, Philip
ABSTRACT This paper will explore the opportunities for artificial intelligence (AI) in the system engineering domain, particularly in ways that unite the unique capabilities of the systems engineer with the AI. This collaboration of human and machine intelligence is known as Augmented Intelligence (AuI). There is little doubt that systems engineering productivity could be improved with effective utilization of well-established AI techniques, such as machine learning, natural language processing, and statistical models. However, human engineers excel at many tasks that remain difficult for AIs, such as visual interpretation, abstract pattern matching, and drawing broad inferences based on experience. Combining the best of AI and human capabilities, along with effective human/machine interactions and data visualization, offers the potential for orders-of-magnitude improvements in the speed and quality of delivered
Petrotta, MarkPeterson, Troy
ABSTRACT Crowdsourcing is an overarching term that denotes a number of ways to use the web as means to enlist a large number of individuals to perform a particular task. The tasks can range from simply providing an opinion, to contributing material, to solving a problem. Because the term crowdsourcing is used to denote a variety of activities in many different contexts, strong opinions have formed in many minds. This paper is an attempt to inform the reader of the complexity that underlies the simple term “crowdsourcing.” We then describe the connection between the DARPA Adaptive Vehicle Make program with the potential limitations of crowdsourcing complex tasks using examples from industry. Using these examples, we present a research motivation detailing areas to be improved within current crowdsourcing frameworks. Finally, an agent-based simulation using machine learning techniques is defined, preliminary results are presented, and future research directions are described
Gerth, Richard J.Burnap, AlexPapalambros, Panos
ABSTRACT Main Battle Tanks (MBTs) remain a key component of most modern militaries. While the best way to ‘kill a tank’ is via the employment of another tank, matching enemy armor formations one-for one is not always possible. Light infantry lack organic armor and their shoulder launched anti-tank capabilities do not defeat the latest generation of MBTs. To compensate for this capability gap, the U.S. Army has employed precision guided anti-tank munitions, such as the “Javelin.” However, these are expensive to produce in quantity and require risking the forward presence of dismounted Soldiers to employ. Mine fields offer another option but are immobile once employed. The ‘Guillotine’ Attack System proposes to change the equation by pairing an AI enabled, adaptive unmanned delivery system with a shaped charge payload. Guillotine can loiter for hours, reposition itself to hunt for targets, and- when ready- deliver a precision shaped charge strike from the air. Citation: “The ‘Guillotine
Dooley, MatthewLacaze, Alberto
ABSTRACT Modern robotic technologies enable the development of semiautonomous ground robots capable of supporting military field operations. Particular attention has been devoted to the robotic mule concept, which aids soldiers in transporting loads over rugged terrain. While existing mule concepts are promising, current configurations are rated for payloads exceeding 1000 lbs., placing them in the size and weight class of small cars and ATVs. These large robots are conspicuous by nature and may not successfully carry out infantry resupply missions in an active combat zone. Conversations with soldiers and industry professionals have spotlighted a need for a compact, lightweight, and low-cost robotic mule. This platform would ensure reliable last-mile delivery of critical supplies to predetermined rally points. We present a design for such a compact robotic mule, the µSMET. This versatile platform will be integrated with the Squad Multipurpose Equipment Transport (SMET), to ferry
Grenn, KatharinaAdam, CristianKleinow, TimothyMason, BrianSapunkov, OlegMuench, PaulLakshmanan, Sridhar
Items per page:
1 – 50 of 1811