Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “detect and avoid”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,423 records · Page 79

A deep learning-based direct forecasting of CO 2 plume migration

Accurate and timely forecasts of CO 2 plume evolution in geological reservoirs are crucial for CO 2 migration detection, leakage risk assessment, and operation decision support. Conventional forecasting usually adopts a two-step strategy, first calibrating reservoir model parameters against observations using iterative inverse modeling (or history matching) and then applying the calibrated model for predictions. This method impedes real-time forecasts due to the heavy computational demand in inverse modeling and may suffer from poor prediction accuracy because of the limited observation data. In this work, we propose a deep learning-based latent space mapping framework to forecast CO 2 plume migration directly by avoiding the inverse modeling. We first use the convolutional autoencoder to map the high-dimensional complex plume extents onto low-dimensional latent space. Next, we use neural networks to learn the relationship between the observation variables and the prediction latent variables. And then for given observation data, we infer the prediction values directly. This one-step direct forecasting is computationally efficient which requires a few number of parallelizable reservoir simulations and it can provide accurate predictions with limited observations by learning the observation-prediction relationship in the reduced dimension. Therefore, our proposed method enables an in-time forecast of dynamic CO 2 plume distributions. In this work, we demonstrate the effectiveness and accuracy of our method in predicting the CO 2 plume migration using four metrics such as plume area, centroid movement distance, and plume spreading in the primary and secondary directions. And the spatio-temporal evolution patterns of plume migration under diverse geological complexities are also accurately quantified.

15 GEOTHERMAL ENERGY↗

Effect of Dosimetric Factors on Occurrence and Volume of Temporal Lobe Necrosis Following Intensity Modulated Radiation Therapy for Nasopharyngeal Carcinoma: A Case-Control Study

Purpose: To determine dosimetric risk factors for the occurrence of temporal lobe necrosis (TLN) among nasopharyngeal carcinoma (NPC) patients treated with intensity modulated radiation therapy (IMRT) and to investigate the impact of dose-volume histogram (DVH) parameters on the volume of TLN lesions (V-N). Methods and Materials: Forty-three NPC patients who had developed TLN following IMRT and 43 control subjects free of TLN were retrospectively assessed. DVH parameters included maximum dose (Dmax), minimum dose (Dmin), mean dose (Dmean), absolute volumes receiving specific dose (Vds) from 20 to 76 Gy (V20-V76), and doses covering certain volumes (Dvs) from 0.25 to 6.0 cm{sup 3} (D0.25-D6.0). V-Ns were quantified with axial magnetic resonance images. Results: DVH parameters were ubiquitously higher in temporal lobes with necrosis than in healthy temporal lobes. Increased Vds and Dvs were significantly associated with higher risk of TLN occurrence (P<.05). In particular, Vds at a dose of ≥70 Gy were found with the highest odds ratios. A common increasing trend was detected between V-N and DVH parameters through trend tests (P for trend of <.05). Linear regression analysis showed that V45 had the strongest predictive power for V-N (adjusted R{sup 2} = 0.305, P<.0001). V45 of <15.1 cm{sup 3} was relatively safe as the dose constraint for preventing large TLN lesions with V-N of >5 cm{sup 3}. Conclusions: Dosimetric parameters are significantly associated with TLN occurrence and the extent of temporal lobe injury. To better manage TLN, it would be important to avoid both focal high dose and moderate dose delivered to a large area in TLs.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Earthquake Magnitude With DAS: A Transferable Data–Based Scaling Relation

Distributed Acoustic Sensing (DAS) is a promising technique to improve the rapid detection and characterization of earthquakes. Previous DAS studies mainly focus on the phase information but less on the amplitude information. In this study, we compile earthquake data from two DAS arrays in California, USA, and one submarine array in Sanriku, Japan. We develop a data-driven method to obtain the first scaling relation between DAS amplitude and earthquake magnitude. Our results reveal that the earthquake amplitudes recorded by DAS in different regions follow a similar scaling relation. The scaling relation can provide a rapid earthquake magnitude estimation and effectively avoid uncertainties caused by the conversion to ground motions. Our results show that the scaling relation appears transferable to new regions with calibrations. The scaling relation highlights the great potential of DAS in earthquake source characterization and early warning.

58 GEOSCIENCES↗

Multiple-camera/motion stereoscopy for range estimation in helicopter flight

Aiding the pilot to improve safety and reduce pilot workload by detecting obstacles and planning obstacle-free flight paths during low-altitude helicopter flight is desirable. Computer vision techniques provide an attractive method of obstacle detection and range estimation for objects within a large field of view ahead of the helicopter. Previous research has had considerable success by using an image sequence from a single moving camera to solving this problem. The major limitations of single camera approaches are that no range information can be obtained near the instantaneous direction of motion or in the absence of motion. These limitations can be overcome through the use of multiple cameras. This paper presents a hybrid motion/stereo algorithm which allows range refinement through recursive range estimation while avoiding loss of range information in the direction of travel. A feature-based approach is used to track objects between image frames. An extended Kalman filter combines knowledge of the camera motion and measurements of a feature's image location to recursively estimate the feature's range and to predict its location in future images. Performance of the algorithm will be illustrated using an image sequence, motion information, and independent range measurements from a low-altitude helicopter flight experiment.

Smith, Phillip N.↗

A communication-avoiding, hybrid-parallel, rank-revealing orthogonalization method.

Orthogonalization consumes much of the run time of many iterative methods for solving sparse linear systems and eigenvalue problems. Commonly used algorithms, such as variants of Gram-Schmidt or Householder QR, have performance dominated by communication. Here, 'communication' includes both data movement between the CPU and memory, and messages between processors in parallel. Our Tall Skinny QR (TSQR) family of algorithms requires asymptotically fewer messages between processors and data movement between CPU and memory than typical orthogonalization methods, yet achieves the same accuracy as Householder QR factorization. Furthermore, in block orthogonalizations, TSQR is faster and more accurate than existing approaches for orthogonalizing the vectors within each block ('normalization'). TSQR's rank-revealing capability also makes it useful for detecting deflation in block iterative methods, for which existing approaches sacrifice performance, accuracy, or both. We have implemented a version of TSQR that exploits both distributed-memory and shared-memory parallelism, and supports real and complex arithmetic. Our implementation is optimized for the case of orthogonalizing a small number (5-20) of very long vectors. The shared-memory parallel component uses Intel's Threading Building Blocks, though its modular design supports other shared-memory programming models as well, including computation on the GPU. Our implementation achieves speedups of 2 times or more over competing orthogonalizations. It is available now in the development branch of the Trilinos software package, and will be included in the 10.8 release.

99 GENERAL AND MISCELLANEOUS↗

Can 18-FDG-PET During Radiotherapy Replace Post-Therapy Scanning for Detection/Demonstration of Tumor Response in Head-and-Neck Cancer?

Purpose: In routine practice, the tumor response in head-and-neck cancer (HNC) is assessed 3-4 months after radiotherapy (RT). We compared the results of fluorodeoxyglucose-positron emission tomography (FDG-PET) during (47 Gy) and 4 months after RT. Methods and Materials: In 40 patients with HNC, PET was performed before (PET1), at the end of Week 4 (47 Gy) (PET2), and 4 months after RT (PET3). Visual analysis classified patients as having a complete response (CR) or a non-CR (NCR). The sensitivity, specificity, accuracy, negative predictive value, and positive predictive value for PET2 and PET3 were determined. The 2-year overall survival (OS) rate for a CR and NCR was calculated for both response evaluation points. Results: After a median follow-up of 26 months, 10 patients had died, 6 had residual disease, and 24 remained disease free. The overall sensitivity, specificity, negative predictive value, positive predictive value, and accuracy of PET2 vs. PET3 for the detection of a CR was 28.6%, 81.8%, 31.0%, 80.0%, and 42.5% vs. 78.6%, 75.0%, 60.0%, 88.0%, and 77.5%, respectively. The 2-year OS rate determined at 47 Gy was 90.0% and 71.8% for a CR and NCR, respectively, and did not appear to be significantly different (p = .50). For the study, at 4 months, the OS was significantly better in the CR group (91.8%) than in the NCR group (49.9%; p = .0055). Conclusion: The high specificity and positive predictive value for the evaluation of tumor response with PET2 and PET3 might avoid unnecessary salvage surgery in patients with a CR. In contrast to PET3, the sensitivity of PET 2 was low, and the difference in OS between the CR and NCR groups was not significantly different. Therefore, the evaluation of the tumor response with FDG-PET at 4 months after RT completion cannot be replaced by FDG-PET during RT at 47 Gy.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Large Game Distribution Camera Study at Los Alamos National Laboratory [Poster]

The unique habitats at the Laboratory support a diverse community of wildlife. The habitat types that include Pinon Juniper, Ponderosa Pine, and Mixed-conifer forests each provide different resources important for a variety of species. Wildlife monitoring for large game animals can be challenging because they avoid interactions with humans, and prefer undeveloped areas. This study was conducted to analyze the distribution of predators and other game animals seasonally across the Laboratory. The data gathered will be used to make recommendations regarding wildlife management for conservation and protection. Utilizing game cameras to remotely monitor wildlife was determined to be the most effective, noninvasive method to gather the necessary data. As camera technology has advanced, game cameras have become adequate for documenting individuals, and are commonly used in wildlife studies on predation, abundance, occupancy, diversity, and endangered species detection. Twenty cameras were placed across the LANL landscape using a modified systematic sampling design. They were deployed in February 2018 and retrieved in January 2019, thus each camera sampled for one full year. All images were reviewed and the species, number, age, and sex of all animals in the photo were recorded. Here we present data for the most abundant species encountered: deer, elk, coyote, bear, bobcat, mountain lion, and fox respectively. The data were summarized by species and season. Future work will include occupancy modelling for each species. This will allow the development of heat maps showing species distributions across the Laboratory.

54 ENVIRONMENTAL SCIENCES↗

Detection of signature consistent with cosmological time dilation in gamma-ray bursts

If gamma ray bursts are at cosmological distances-as suggested by their isotropic distribution on the sky and by their number-intensity relation-then the burst profiles will be stretched in time, by an amount proportional to the redshift, 1 + Z. We have tested data from the Compton Gamma Ray Observatory's (CGRO's) Burst and Transient Source Experiment (BATSE) for such time dilation. Out of 590 bursts observed by BATSE, 131 bursts were analyzed; bursts with durations shorter than 1.5 s were excluded. We used three tests to compare the timescales of bright and dim bursts, the latter, on average, being more distant than the former. Our measures of timescale are constructed to avoid selection effects arising from intensity differences by rescaling all bursts to fiducial levels of peak intensity and noise bias. (1) We found that the total rescaled count above background for the dim burst ensemble is approximately twice that for the brightest bursts-translating into longer durations for the dim bursts. (2) Wavelet-transform decompositions of the burst profiles confirmed that this dilation operates over a broad range of timescales. (3) Structure on the shortest timescales was examined using a procedure which aligns the highest peaks of profiles from which the noise has been optimally removed using a wavelet threshold technique. In all three tests, the dim bursts are stretched by a factor of approximately 2 relative to the bright ones, over seven octaves of timescale. We calibrated the measurements by dilating synthetic bursts that approximate the temporal characteristics of bright BATSE bursts. Results are consistent with bursts at BATSE's peak-flux completeness limit being at cosmological distances corresponding to Z approximately equal to 1, and thus with independent cosmological interpretations of the BATSE number-intensity relation. Alternative explanations of our results, arising from the nature of physical processes in bursts, are still possible.

Norris, J. P.↗

Gravitropic responses of the Avena coleoptile in space and on clinostats. II. Is reciprocity valid?

Experiments were undertaken to determine if the reciprocity rule is valid for gravitropic responses of oat coleoptiles in the acceleration region below 1 g. The rule predicts that the gravitropic response should be proportional to the product of the applied acceleration and the stimulation time. Seedlings were cultivated on 1 g centrifuges and transferred to test centrifuges to apply a transverse g-stimulation. Since responses occurred in microgravity, the uncertainties about the validity of clinostat simulation of weightlessness was avoided. Plants at two stages of coleoptile development were tested. Plant responses were obtained using time-lapse video recordings that were analyzed after the flight. Stimulus intensities and durations were varied and ranged from 0.1 to 1.0 g and from 2 to 130 min, respectively. For threshold g-doses the reciprocity rule was obeyed. The threshold dose was of the order of 55 g s and 120 g s, respectively, for two groups of plants investigated. Reciprocity was studied also at bending responses which are from just above the detectable level to about 10 degrees. The validity of the rule could not be confirmed for higher g-doses, chiefly because the data were more variable. It was investigated whether the uniformity of the overall response data increased when the gravitropic dose was defined as (gm x t) with m-values different from unity. This was not the case and the reciprocity concept is, therefore, valid also in the hypogravity region. The concept of gravitropic dose, the product of the transverse acceleration and the stimulation time, is also well-defined in the acceleration region studied. With the same hardware, tests were done on earth where responses occurred on clinostats. The results did not contradict the reciprocity rule but scatter in the data was large.

Flight Experiment↗

Stereo-Optic High Definition Imaging: A New Technology to Understand Bird and Bat Avoidance of Wind Turbines

There is a critical need to develop monitoring tools to track aerofauna (birds and bats) in three dimensions around wind turbines. New monitoring systems will reduce permitting uncertainty by increasing the understanding of how birds and bats are interacting with wind turbines, which will improve the accuracy of impact predictions. Biodiversity Research Institute (BRI), The University of Maine Orono School of Computing and Information Science (UMaine SCIS), HiDef Aerial Surveying Limited (HiDef), and SunEdison, Inc. (formerly First Wind) responded to this need by using stereo-optic cameras with near-infrared (nIR) technology to investigate new methods for documenting aerofauna behavior around wind turbines. The stereo-optic camera system used two synchronized high-definition video cameras with fisheye lenses and processing software that detected moving objects, which could be identified in post-processing. The stereo- optic imaging system offered the ability to extract 3-D position information from pairs of images captured from different viewpoints. Fisheye lenses allowed for a greater field of view, but required more complex image rectification to contend with fisheye distortion. The ability to obtain 3-D positions provided crucial data on the trajectory (speed and direction) of a target, which, when the technology is fully developed, will provide data on how animals are responding to and interacting with wind turbines. This project was focused on testing the performance of the camera system, improving video review processing time, advancing the 3-D tracking technology, and moving the system from Technology Readiness Level 4 to 5. To achieve these objectives, we determined the size and distance at which aerofauna (particularly eagles) could be detected and identified, created efficient data management systems, improved the video post-processing viewer, and attempted refinement of 3-D modeling with respect to fisheye lenses. The 29-megapixel camera system successfully captured 16,173 five-minute video segments in the field. During nighttime field trials using nIR, we found that bat-sized objects could not be detected more than 60 m from the camera system. This led to a decision to focus research efforts exclusively on daytime monitoring and to redirect resources towards improving the video post- processing viewer. We redesigned the bird event post-processing viewer, which substantially decreased the review time necessary to detect and identify flying objects. During daytime field trials, we determine that eagles could be detected up to 500 m away using the fisheye wide-angle lenses, and eagle-sized targets could be identified to species within 350 m of the camera system. We used distance sampling survey methods to describe the probability of detecting and identifying eagles and other aerofauna as a function of distance from the system. The previously developed 3-D algorithm for object isolation and tracking was tested, but the image rectification (flattening) required to obtain accurate distance measurements with fish-eye lenses was determined to be insufficient for distant eagles. We used MATLAB and OpenCV to improve fisheye lens rectification towards the center of the image, but accurate measurements towards the image corners could not be achieved. We believe that changing the fisheye lens to rectilinear lens would greatly improve position estimation, but doing so would result in a decrease in viewing angle and depth of field. Finally, we generated simplified shape profiles of birds to look for similarities between unknown animals and known species. With further development, this method could provide a mechanism for filtering large numbers of shapes to reduce data storage and processing. These advancements further refined the camera system and brought this new technology closer to market. Once commercialized, the stereo-optic camera system technology could be used to: a) research how different species interact with wind turbines in order to refine collision risk models and inform mitigation solutions; and b) monitor aerofauna interactions with terrestrial and offshore wind farms replacing costly human observers and allowing for long-term monitoring in the offshore environment. The camera system will provide developers and regulators with data on the risk that wind turbines present to aerofauna, which will reduce uncertainty in the environmental permitting process.

Adams, Evan↗

Adaptive Fault Current-Limiting Control of MMC for Protection of Multiterminal HVDC Systems

A crucial requirement of the protection system for multi-terminal high-voltage DC (MTDC) transmission is that it is capable of selectively isolating the faulty area from the healthy part of the network using DC circuit breakers (DCCBs), while ensuring continuous operation of converter stations in the healthy part of the network. But in the half-bridge modular multilevel converters (HB-MMCs) based MTDC system, since HB-MMCs do not have fault current absorption capability, when a DC fault occurs, the rising fault currents can quickly reach the blocking threshold within a few milliseconds and disrupt the operation of the MMCs in the healthy part. To facilitate fault-ride-through capability of MTDC system, large DC reactors are often considered in series with DCCBs to reduce the rate of rise of the fault current and prevent blocking the MMCs in the healthy part of the DC networks; however, large DC reactors prohibitively increase the cost of the system, introduce stability issues and can create post-fault oscillations. This paper presents an adaptive fault current-limiting control method for MMCs to avoid their blocking and enable the continuous operation of the healthy part of MTDC systems. The method contains two parts: The first part is based on circulating current feedforward control, which emulates virtual reactors in each arm of an MMC, and is immediately activated when the fault current starts to increase to reduce the rate of rise of the fault current. The second part temporarily bypasses all the submodules when the fault current exceeds a preset threshold, complementing the fault current-limiting effect of the first part. Neither part requires fault detection signals, and they are automatically activated during faults. Simulation case studies of a four-terminal bipolar MMC-based high-voltage DC system are presented to demonstrate the effectiveness of the proposed control methods.

active fault current limiting↗

RELICS OF GALAXY MERGING: OBSERVATIONAL PREDICTIONS FOR A WANDERING MASSIVE BLACK HOLE AND ACCOMPANYING STAR CLUSTER IN THE HALO OF M31

Galaxies and massive black holes (BHs) presumably grow via galactic merging events and subsequent BH coalescence. As a case study, we investigate the merging event between the Andromeda galaxy (M31) and a satellite galaxy. We compute the expected observational appearance of the massive BH that was at the center of the satellite galaxy prior to the merger and is currently wandering in the M31 halo. We demonstrate that a radiatively inefficient accretion flow with a bolometric luminosity of a few tens of solar luminosities develops when Hoyle-Lyttleton accretion onto the BH is assumed. We compute the associated broadband spectrum and show that the radio band (observable with EVLA, ALMA, and the Square Kilometre Array) is the best frequency range in which to detect the emission. We also evaluate the mass and the luminosity of the stars bound by the wandering BH and find that such a star cluster is sufficiently luminous that it could correspond to one of the star clusters found by the PAndAS survey. The discovery of a relic massive BH wandering in a galactic halo will provide a direct means of investigating in detail the coevolution of galaxies and BHs. It also means a new population of BHs (off-center massive BHs) and offers targets for clean BH imaging that avoid strong interstellar scattering in the centers of galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Autonomous System Subversion Tactics: Prototypes and Recommended Countermeasures

One of the fielding requirements for Advanced and Small Modular Reactors (AR/SMR) is the ability to support remote and autonomous operations. Autonomous Control Systems (ACS) are found on platforms such as Autonomous Space Vehicles, Cruise Missiles, and advanced driver-assistance systems. Each of these ACS implementations depends upon a set of decision support subsystems responsible for supporting Autonomous Mission Managers (names vary based upon field and author preferences). These Autonomous Mission Managers receive inputs from system sensors (e.g., LIDAR collection from an automobile travelling down a street; transients from a nuclear reactor), and perform a set of classifications (e.g., Red Traffic Light; Small Pedestrian at 10m; Load Rejection; Single Coolant Pump Trip), and then use these classifications in combination with recommendation algorithms to achieve platform goals (e.g., Stop the Vehicle at the Traffic Light, Avoid the Small Pedestrian; Trip the Reactor to prevent a Safety Event). The design, implementation, and fielding of an ACS capability will alter the cyber-attack surface such that existing risk management plans will need to be updated to include how to protect and defend against data-science and decision-support-system attack classes. These attack classes would include protection of the design and training environments where algorithm selection and testing and training data would be obvious attack vectors. These attack classes would also require an informed set of detection and response procedures to identify anomalous behaviors and document best practices for anomaly assessment and vulnerability mitigation and remediation. Last year we published a Cyber Threat Assessment Methodology for Autonomous and Remote Operations for AR/SMRs along with a companion publication on Cyber Attack and Defense Use Cases. The focus of the methodology was on describing and enumerating ACS processes, components, and functions such that security engineers could: evaluate subversion options against the target; identify threat actor attributes and capabilities derived from each subversion option; and identify security controls and response countermeasures. The Use Cases document offered detailed methodology examples including an assessment of a Military Base SMR, an Autonomous System Decision Loop, and implementation of AR/SMR Machine Learning algorithms. Our proposal at the end of last year was to focus on implementation of subversion prototypes related to the last Use Case area: AR/SMR Machine Learning (ML) Algorithms. We included six attack scenarios in our Use Cases paper: a Poisoning Attack against ML functions implemented using an FPGA; a Trojaning Attack against ML classifiers exploiting the excitability of Nuclear Engineers; a Backdooring Attack against ML Training environments to ensure persistence of an attack vector; a False Positive Evasion Attack against multi-factor Access Control Systems using clever inputs; an Inference Attack against ML models by an Insider with access to the Operational environment; and an Adversarial Reprogramming Attack against a Material Access Control Video Surveillance System. At the beginning of this year these six attack scenarios were provided to our research teams at Georgia Tech and Idaho State University and each team successfully implemented a subversion attack against a ML implementation to include transient misclassifications. While this is a notable outcome from this type of research, this paper offers the reader insight into not only how to structure and execute these types of attacks, but into the thought process behind how the researcher investigated the problem space, performed initial algorithm implementation, and the trial-and-error behind arriving at the successful subversion prototypes. We include in this paper a set of associated Scenarios on how these subversion prototypes could be implemented and an initial set of guidance for AR/SMR architects, Nuclear Regulators, and Cyber Defenders to implement awareness and defense capabilities into their current operational portfolios.

42 ENGINEERING↗

The Highly Robust Electrical Interconnects and Ultrasensitive Biosensors Based on Embedded Carbon Nanotube Arrays

We report on our recent breakthroughs in two different applications using well-aligned carbon nanotube (CNT) arrays on Si chips, including (1) a novel processing solution for highly robust electrical interconnects in integrated circuit manufacturing, and (2) the development of ultrasensitive electrochemical DNA sensors. Both of them rely on the invention of a bottom-up fabrication scheme which includes six steps, including: (a) lithographic patterning, (b) depositing bottom conducting contacts, (c) depositing metal catalysts, (d) CNT growth by plasma enhanced chemical vapor deposition (PECVD), (e) dielectric gap-filling, and (f) chemical mechanical polishing (CMP). Such processes produce a stable planarized surface with only the open end of CNTs exposed, whch can be further processed or modified for different applications. By depositing patterned top contacts, the CNT can serve as vertical interconnects between the two conducting layers. This method is fundamentally different fiom current damascene processes and avoids problems associated with etching and filling of high aspect ratio holes at nanoscales. In addition, multiwalled CNTs (MWCNTs) are highly robust and can carry a current density of 10(exp 9) A/square centimeters without degradation. It has great potential to help extending the current Si technology. The embedded MWCNT array without the top contact layer can be also used as a nanoelectrode array in electrochemical biosensors. The cell time-constant and sensitivity can be dramatically improved. By functionalizing the tube ends with specific oligonucleotide probes, specific DNA targets can be detected with electrochemical methods down to subattomoles.

Li, Jun↗

Adaptive Liver Stereotactic Body Radiation Therapy: Automated Daily Plan Reoptimization Prevents Dose Delivery Degradation Caused by Anatomy Deformations

Purpose: To investigate how dose distributions for liver stereotactic body radiation therapy (SBRT) can be improved by using automated, daily plan reoptimization to account for anatomy deformations, compared with setup corrections only. Methods and Materials: For 12 tumors, 3 strategies for dose delivery were simulated. In the first strategy, computed tomography scans made before each treatment fraction were used only for patient repositioning before dose delivery for correction of detected tumor setup errors. In adaptive second and third strategies, in addition to the isocenter shift, intensity modulated radiation therapy beam profiles were reoptimized or both intensity profiles and beam orientations were reoptimized, respectively. All optimizations were performed with a recently published algorithm for automated, multicriteria optimization of both beam profiles and beam angles. Results: In 6 of 12 cases, violations of organs at risk (ie, heart, stomach, kidney) constraints of 1 to 6 Gy in single fractions occurred in cases of tumor repositioning only. By using the adaptive strategies, these could be avoided (<1 Gy). For 1 case, this needed adaptation by slightly underdosing the planning target volume. For 2 cases with restricted tumor dose in the planning phase to avoid organ-at-risk constraint violations, fraction doses could be increased by 1 and 2 Gy because of more favorable anatomy. Daily reoptimization of both beam profiles and beam angles (third strategy) performed slightly better than reoptimization of profiles only, but the latter required only a few minutes of computation time, whereas full reoptimization took several hours. Conclusions: This simulation study demonstrated that replanning based on daily acquired computed tomography scans can improve liver stereotactic body radiation therapy dose delivery.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Towards a CAN IDS Based on a Neural Network Data Field Predictor

Modern vehicles contain a few controller area networks (CANs), which allow scores of on-board electronic control units (ECUs) to communicate messages critical to vehicle functions and driver safety. CAN provides a lightweight and reliable broadcast protocol but is bereft of security features. As evidenced by many recent research works, CAN exploits are possible both remotely and with direct access, fueling a growing CAN intrusion detection system (IDS) body of research. A challenge for pioneering vehicle-agnostic IDSs is that passenger vehicles' CAN message encodings are proprietary, defined and held secret by original equipment manufacturers (OEMs). Targeting detection of next-generation attacks, in which messages are sent from the expected ECU at the expected time but with malicious content, researchers are now seeking to leverage "CAN data models'', which predict future CAN messages and use prediction error to identify anomalous, hopefully malicious CAN messages. Yet, current works model CAN signals post-translation, i.e., after applying OEM-donated or reverse-engineered translations from raw data. We present initial IDS results testing deep neural networks used to predict CAN data at the bit level, targeting IDS capabilities that avoiding reverse engineering proprietary encodings. Our results suggest the method is promising for data with signals exhibiting dependence on previous or concurrent inputs.

Pawelec, Krzysztof↗

Second Line of Defense Program, Secondary Screening Operations Field Data Collection

Secondary screening is a key component of the Second Line of Defense (SLD) Program’s capability to deter, detect, and interdict the illicit trafficking of radioactive materials. The primary detection system—the radiation portal monitor—makes alarm decisions based on gross counts of gamma and neutron radiation; it is able to provide little other information regarding what may have caused an elevated radiation level. Conducting a proper secondary screening with an instrument capable of identifying radioisotopes is the only way to determine what radioactive material is present—most importantly if it is a threat material. Only by knowing the radioactive material can a prudent decision be made with regard to disposition and response—and can the mistaken release of a threat be avoided.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Stereoscopic Machine-Vision System Using Projected Circles

A machine-vision system capable of detecting obstacles large enough to damage or trap a robotic vehicle is undergoing development. The system includes (1) a pattern generator that projects concentric circles of laser light forward onto the terrain, (2) a stereoscopic pair of cameras that are aimed forward to acquire images of the circles, (3) a frame grabber and digitizer for acquiring image data from the cameras, and (4) a single-board computer that processes the data. The system is being developed as a prototype of machine- vision systems to enable robotic vehicles ( rovers ) on remote planets to avoid craters, large rocks, and other terrain features that could capture or damage the vehicles. Potential terrestrial applications of systems like this one could include terrain mapping, collision avoidance, navigation of robotic vehicles, mining, and robotic rescue. This system is based partly on the same principles as those of a prior stereoscopic machine-vision system in which the cameras acquire images of a single stripe of laser light that is swept forward across the terrain. However, this system is designed to afford improvements over some of the undesirable features of the prior system, including the need for a pan-and-tilt mechanism to aim the laser to generate the swept stripe, ambiguities in interpretation of the single-stripe image, the time needed to sweep the stripe across the terrain and process the data from many images acquired during that time, and difficulty of calibration because of the narrowness of the stripe. In this system, the pattern generator does not contain any moving parts and need not be mounted on a pan-and-tilt mechanism: the pattern of concentric circles is projected steadily in the forward direction. The system calibrates itself by use of data acquired during projection of the concentric-circle pattern onto a known target representing flat ground. The calibration- target image data are stored in the computer memory for use as a template in processing terrain images. During operation on terrain, the images acquired by the left and right cameras are analyzed. The analysis includes (1) computation of the horizontal and vertical dimensions and the aspect ratios of rectangles that bound the circle images and (2) comparison of these aspect ratios with those of the template. Coordinates of distortions of the circles are used to identify and locate objects. If the analysis leads to identification of an object of significant size, then stereoscopicvision algorithms are used to estimate the distance to the object. The time taken in performing this analysis on a single pair of images acquired by the left and right cameras in this system is a fraction of the time taken in processing the many pairs of images acquired in a sweep of the laser stripe across the field of view in the prior system. The results of the analysis include data on sizes and shapes of, and distances and directions to, objects. Coordinates of objects are updated as the vehicle moves so that intelligent decisions regarding speed and direction can be made. The results of the analysis are utilized in a computational decision-making process that generates obstacle-avoidance data and feeds those data to the control system of the robotic vehicle.

Mackey, Jeffrey R.↗