Performance Portability in the Uintah Runtime System Through the Use of Kokkos.
Abstract not provided.
SEARCH · Search NASA
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.
Abstract not provided.
Purpose: To evaluate setup error and interfractional changes in tumor motion magnitude using an electric portal imaging device in cine mode (EPID cine) during the course of stereotactic body radiation therapy (SBRT) for non-small-cell lung cancer (NSCLC) and to calculate margins to compensate for these variations. Materials and Methods: Subjects were 28 patients with Stage I NSCLC who underwent SBRT. Respiratory-correlated four-dimensional computed tomography (4D-CT) at simulation was binned into 10 respiratory phases, which provided average intensity projection CT data sets (AIP). On 4D-CT, peak-to-peak motion of the tumor (M-4DCT) in the craniocaudal direction was assessed and the tumor center (mean tumor position [MTP]) of the AIP (MTP-4DCT) was determined. At treatment, the tumor on cone beam CT was registered to that on AIP for patient setup. During three sessions of irradiation, peak-to-peak motion of the tumor (M-cine) and the mean tumor position (MTP-cine) were obtained using EPID cine and in-house software. Based on changes in tumor motion magnitude ( Increment M) and patient setup error ( Increment MTP), defined as differences between M-4DCT and M-cine and between MTP-4DCT and MTP-cine, a margin to compensate for these variations was calculated with Stroom's formula. Results: The means ({+-}standard deviation: SD) of M-4DCT and M-cine were 3.1 ({+-}3.4) and 4.0 ({+-}3.6) mm, respectively. The means ({+-}SD) of Increment M and Increment MTP were 0.9 ({+-}1.3) and 0.2 ({+-}2.4) mm, respectively. Internal target volume-planning target volume (ITV-PTV) margins to compensate for Increment M, Increment MTP, and both combined were 3.7, 5.2, and 6.4 mm, respectively. Conclusion: EPID cine is a useful modality for assessing interfractional variations of tumor motion. The ITV-PTV margins to compensate for these variations can be calculated.
This paper proposes a fundamental frequency sorting algorithm for balancing the floating capacitors in modular multilevel converters with low-frequency carrier phase shift modulation. The relationship between the driving pulses and submodules (SMs) is rebuilt in every fundamental period by sorting the voltage increments and present voltages. Besides, an improved and simplified strategy is proposed to predict the charging abilities of the driving pulses, so as to avoid sorting the voltage increments. Based on the proposed method, the switching frequency of each power device and the carrier frequency are identical, which reduces the operational losses for applications regarding high power. At the same time, the sorting frequency is as low as the fundamental frequency, which alleviates a large amount of computational cost, and the necessity to measure arm currents is omitted. In this case, the arm current sensors are eliminated, and simplified communication among the central controller and the local ones is set up. Finally, a three-phase 1-MVA simulation platform with 120 SMs and a down-scaled 6-kVA experimental prototype with 48 SMs are constructed to validate the proposed approach.
The Los Alamos sea ice model (CICE) is the result of an effort to develop a computationally efficient sea ice component for a fully coupled atmosphere–land–ocean–ice global climate model. It was originally designed to be compatible with the Parallel Ocean Program (POP), an ocean circulation model developed at Los Alamos National Laboratory for use on massively parallel computers. CICE has several interacting components: a vertical thermodynamic model that computes local growth rates of snow and ice due to vertical conductive, radiative and turbulent fluxes, along with snowfall; an elastic-viscous-plastic model of ice dynamics, which predicts the velocity field of the ice pack based on a model of the material strength of the ice; an incremental remapping transport model that describes horizontal advection of the areal concentration, ice and snow volume and other state variables; and a ridging parameterization that transfers ice among thickness categories based on energetic balances and rates of strain. It also includes a biogeochemical model that describes evolution of the ice ecosystem. The CICE sea ice model is used for climate research as one component of complex global earth system models that include atmosphere, land, ocean and biogeochemistry components. It is also used for operational sea ice forecasting in the polar regions and in numerical weather prediction models.
The Los Alamos sea ice model (CICE) is the result of an effort to develop a computationally efficient sea ice component for a fully coupled atmosphere–land–ocean–ice global climate model. It was originally designed to be compatible with the Parallel Ocean Program (POP), an ocean circulation model developed at Los Alamos National Laboratory for use on massively parallel computers. CICE has several interacting components: a vertical thermodynamic model that computes local growth rates of snow and ice due to vertical conductive, radiative and turbulent fluxes, along with snowfall; an elastic-viscous-plastic model of ice dynamics, which predicts the velocity field of the ice pack based on a model of the material strength of the ice; an incremental remapping transport model that describes horizontal advection of the areal concentration, ice and snow volume and other state variables; and a ridging parameterization that transfers ice among thickness categories based on energetic balances and rates of strain. It also includes a biogeochemical model that describes evolution of the ice ecosystem. The CICE sea ice model is used for climate research as one component of complex global earth system models that include atmosphere, land, ocean and biogeochemistry components. It is also used for operational sea ice forecasting in the polar regions and in numerical weather prediction models.
In the late 1980's and early 1990's, Lawrence Livermore National Laboratory was deeply engrossed in determining the next generation programming model for the Integrated Design Codes (IDC) beyond vectorization for the Cray 1s series of computers. The vector model, developed in mid 1970's first for the CDC 7600 and later extended from stack based vector operation to memory to memory operations for the Cray 1s, lasted approximately 20 years (See Slide 5). The Cray vector era was deemed an extremely long lived era as it allowed vector codes to be developed over time (the Cray 1s were faster in scalar mode than the CDC 7600) with vector unit utilization increasing incrementally over time. The other attributes of the Cray vector era at LLNL were that we developed, supported and maintained the Operating System (LTSS and later NLTSS), communications protocols (LINCS), Compilers (Civic Fortran77 and Model), operating system tools (e.g., batch system, job control scripting, loaders, debuggers, editors, graphics utilities, you name it) and math and highly machine optimized libraries (e.g., SLATEC, and STACKLIB). Although LTSS was adopted by Cray for early system generations, they later developed COS and UNICOS operating systems and environment on their own. In the late 1970s and early 1980s two trends appeared that made the Cray vector programming model (described above including both the hardware and system software aspects) seem potentially dated and slated for major revision. These trends were the appearance of low cost CMOS microprocessors and their attendant, departmental and mini-computers and later workstations and personal computers. With the wide spread adoption of Unix in the early 1980s, it appeared that LLNL (and the other DOE Labs) would be left out of the mainstream of computing without a rapid transition to these 'Killer Micros' and modern OS and tools environments. The other interesting advance in the period is that systems were being developed with multiple 'cores' in them and called Symmetric Multi-Processor or Shared Memory Processor (SMP) systems. The parallel revolution had begun. The Laboratory started a small 'parallel processing project' in 1983 to study the new technology and its application to scientific computing with four people: Tim Axelrod, Pete Eltgroth, Paul Dubois and Mark Seager. Two years later, Eugene Brooks joined the team. This team focused on Unix and 'killer micro' SMPs. Indeed, Eugene Brooks was credited with coming up with the 'Killer Micro' term. After several generations of SMP platforms (e.g., Sequent Balance 8000 with 8 33MHz MC32032s, Allian FX8 with 8 MC68020 and FPGA based Vector Units and finally the BB&N Butterfly with 128 cores), it became apparent to us that the killer micro revolution would indeed take over Crays and that we definitely needed a new programming and systems model. The model developed by Mark Seager and Dale Nielsen focused on both the system aspects (Slide 3) and the code development aspects (Slide 4). Although now succinctly captured in two attached slides, at the time there was tremendous ferment in the research community as to what parallel programming model would emerge, dominate and survive. In addition, we wanted a model that would provide portability between platforms of a single generation but also longevity over multiple--and hopefully--many generations. Only after we developed the 'Livermore Model' and worked it out in considerable detail did it become obvious that what we came up with was the right approach. In a nutshell, the applications programming model of the Livermore Model posited that SMP parallelism would ultimately not scale indefinitely and one would have to bite the bullet and implement MPI parallelism within the Integrated Design Code (IDC). We also had a major emphasis on doing everything in a completely standards based, portable methodology with POSIX/Unix as the target environment. We decided against specialized libraries like STACKLIB for performance, but kept as many general purpose, portable math libraries as were needed by the codes. Third, we assumed that the SMPs in clusters would evolve in time to become more powerful, feature rich and, in particular, offer more cores. Thus, we focused on OpenMP, and POSIX PThreads for programming SMP parallelism. These code porting efforts were lead by Dale Nielsen, A-Division code group leader, and Randy Christensen, B-Division code group leader. Most of the porting effort revolved removing 'Crayisms' in the codes: artifacts of LTSS/NLTSS, Civic compiler extensions beyond Fortran77, IO libraries and dealing with new code control languages (we switched to Perl and later to Python). Adding MPI to the codes was initially problematic and error prone because the programmers used MPI directly and sprinkled the calls throughout the code.
In this paper, we develop a method which we call OnlineGCP for computing the Generalized Canonical Polyadic (GCP) tensor decomposition of streaming data. GCP differs from traditional canonical polyadic (CP) tensor decompositions as it allows for arbitrary objective functions which the CP model attempts to minimize. This approach can provide better fits and more interpretable models when the observed tensor data is strongly non-Gaussian. In the streaming case, tensor data is gradually observed over time and the algorithm must incrementally update a GCP factorization with limited access to prior data. In this work, we extend the GCP formalism to the streaming context by deriving a GCP optimization problem to be solved as new tensor data is observed, formulate a tunable history term to balance reconstruction of recently observed data with data observed in the past, develop a scalable solution strategy based on segregated solves using stochastic gradient descent methods, describe a software implementation that provides performance and portability to contemporary CPU and GPU architectures and integrates with Matlab for enhanced usability, and demonstrate the utility and performance of the approach and software on several synthetic and real tensor data sets.
In ITER and future fusion power plants, the source of the fusion born alpha particles is almost isotropic in pitch angle, thus having co- and counter-current populations. For trapped ions, the co-current side of the orbit corresponds to its outer leg, while the counter-current side corresponds to the inner leg. Understanding the mechanisms responsible for the fast-ion losses (FILs) is critical for future magnetically confined fusion power plants. To further study the interplay of fast ions with plasma instabilities, a double pinhole collimator has been developed for a Fast-Ion Loss Detector (FILD) in the ASDEX Upgrade tokamak (AUG). This new FILD opens the operational window to simultaneous measurements of the co- and counter-current ion velocity-space. In this paper, the first results for the AUG double collimator FILD detector are shown. The commissioning of this new probe is carried out in H-mode plasmas with an on-axis magnetic field $B_0 = -2.5$ T, and a plasma current $I_{\textrm{p}} = 0.7$ MA. Simultaneous co- and counter-current FILs have been measured. Both have shown a similar dependence on Ion Cyclotron Resonance Heating (ICRH) power, where the main difference is the intensity of the losses, with the co-losses being an order of magnitude larger. Toroidal Alfvén eigenmode-coherent ICRH-only losses have been identified for the co-current ions. Additionally, the presence of Edge Localized Modes during the discharge were shown to increment Neutral Beam Injection prompt losses, while partially mitigating ICRH-driven losses on both co- and counter- sides of the velocity-space. Finally, a very trapped and high gyroradius losses, with an unclear origin, have been measured in the co- and counter-current velocity-space. The computed ion trajectories show that these ions remain permanently near the vessel wall, suggesting that they are accelerated within the scrape-off layer.
Purpose: Rigid 2D-3D registration is an alternative to 3D-3D registration for cases where largely bony anatomy can be used for patient positioning in external beam radiation therapy. In this article, the authors evaluated seven similarity measures for use in the intensity-based rigid 2D-3D registration using a variation in Skerl's similarity measure evaluation protocol. Methods: The seven similarity measures are partitioned intensity uniformity, normalized mutual information (NMI), normalized cross correlation (NCC), entropy of the difference image, pattern intensity (PI), gradient correlation (GC), and gradient difference (GD). In contrast to traditional evaluation methods that rely on visual inspection or registration outcomes, the similarity measure evaluation protocol probes the transform parameter space and computes a number of similarity measure properties, which is objective and optimization method independent. The variation in protocol offers an improved property in the quantification of the capture range. The authors used this protocol to investigate the effects of the downsampling ratio, the region of interest, and the method of the digitally reconstructed radiograph (DRR) calculation [i.e., the incremental ray-tracing method implemented on a central processing unit (CPU) or the 3D texture rendering method implemented on a graphics processing unit (GPU)] on the performance of the similarity measures. The studies were carried out using both the kilovoltage (kV) and the megavoltage (MV) images of an anthropomorphic cranial phantom and the MV images of a head-and-neck cancer patient. Results: Both the phantom and the patient studies showed the 2D-3D registration using the GPU-based DRR calculation yielded better robustness, while providing similar accuracy compared to the CPU-based calculation. The phantom study using kV imaging suggested that NCC has the best accuracy and robustness, but its slow function value change near the global maximum requires a stricter termination condition for an optimization method. The phantom study using MV imaging indicated that PI, GD, and GC have the best accuracy, while NCC and NMI have the best robustness. The clinical study using MV imaging showed that NCC and NMI have the best robustness. Conclusions: The authors evaluated the performance of seven similarity measures for use in 2D-3D image registration using the variation in Skerl's similarity measure evaluation protocol. The generalized methodology can be used to select the best similarity measures, determine the optimal or near optimal choice of parameter, and choose the appropriate registration strategy for the end user in his specific registration applications in medical imaging.
ABSTRACT This work presents a micromechanical spectral formulation for obtaining the full‐field and homogenized response of elastoplastic micropolar composites. A closed‐form radial‐return mapping is derived from thermodynamics‐based micropolar elastoplastic constitutive equations to determine the increment of plastic strain necessary to return the generalized stress state to the yield surface, and the algorithm implementation is verified using the method of numerically manufactured solutions. Then, size‐dependent material response and micro‐plasticity are shown as features that may be efficiently simulated in this micropolar elastoplastic framework. The computational efficiency of the formulation enables the generation of large datasets in reasonable computing times.
The vibration characteristic of the sandwich plate with a honeycomb core was investigated by experimental measurements and numerical calculation based on radial basis function (RBF). RBF method was used not only in the meshless approach but also in the post-processing of the experimental data. During the experiment, amplitude-fluctuation electronic speckle pattern interferometry was applied to access the resonant frequencies and the corresponding vibration mode shapes simultaneously. Then RBF method was used to improve the quality of patterns and reconstruct the out-of-plane vibration amplitude after fringe analysis. As for numerical calculation, the modal parameters were numerically predicted using the first-order shear deformation theory. The computation approach was based on collocation with multi-quadric radial basis function. To understand the influence of the thickness of face sheet on dynamic behaviors, three types of specimens with different thickness were tested and analyzed. Of particular interest was that the vibration modes show veering due to the thickness increment. Furthermore, the numerical predicted results were compared with the experimental measurements for the first five modes. They are in good agreement with each other for resonant frequencies, mode shapes and relative out-of-plane amplitudes.
Next-generation exascale systems, those capable of performing a quintillion (10{sup 18}) operations per second, are expected to be delivered in the next 8-10 years. These systems, which will be 1,000 times faster than current systems, will be of unprecedented scale. As these systems continue to grow in size, faults will become increasingly common, even over the course of small calculations. Therefore, issues such as fault tolerance and reliability will limit application scalability. Current techniques to ensure progress across faults like checkpoint/restart, the dominant fault tolerance mechanism for the last 25 years, are increasingly problematic at the scales of future systems due to their excessive overheads. In this work, we evaluate a number of techniques to decrease the overhead of checkpoint/restart and keep this method viable for future exascale systems. More specifically, this work evaluates state-machine replication to dramatically increase the checkpoint interval (the time between successive checkpoint) and hash-based, probabilistic incremental checkpointing using graphics processing units to decrease the checkpoint commit time (the time to save one checkpoint). Using a combination of empirical analysis, modeling, and simulation, we study the costs and benefits of these approaches on a wide range of parameters. These results, which cover of number of high-performance computing capability workloads, different failure distributions, hardware mean time to failures, and I/O bandwidths, show the potential benefits of these techniques for meeting the reliability demands of future exascale platforms.
Graph pattern matching involves finding exact or approximate matches for a query subgraph in a larger graph. It has been studied extensively and has strong applications in domains such as computer vision, computational biology, social networks, security and finance. The problem of exact graph pattern matching is often described in terms of subgraph isomorphism which is NP-complete. The exponential growth in streaming data from online social networks, news and video streams and the continual need for situational awareness motivates a solution for finding patterns in streaming updates. This is also the prime driver for the real-time analytics market. Development of incremental algorithms for graph pattern matching on streaming inputs to a continually evolving graph is a nascent area of research. Some of the challenges associated with this problem are the same as found in continuous query (CQ) evaluation on streaming databases. This paper reviews some of the representative work from the exhaustively researched field of CQ systems and identifies important semantics, constraints and architectural features that are also appropriate for HPC systems performing real-time graph analytics. For each of these features we present a brief discussion of the challenge encountered in the database realm, the approach to the solution and state their relevance in a high-performance, streaming graph processing framework.
Precision machining tool wear is responsible for low product throughput and quality. Monitoring the tool wear online is vital to prevent degradation in machining quality. However, direct real-time tool wear measurement is not practical. This paper presents residual-based anomaly detection models, combining a hybrid model comprised of a physics-based model and a data-driven model (a decision tree or a neural network) to predict signals of interest (e.g., power or forces) under nominal conditions, followed by Page’s cumulative sum test for detecting tool wear on-line using the computer numerical control machine measurements. The most informative features are ranked using dynamic programming and its approximation variants from real-time measurements and machine settings, such as the width of cut, depth of cut, feed rate and spindle speed, that serve as inputs to the predictive models. The baseline nominal model is incrementally updated with experimental data via a gradient boosted adaptation model to generate the residuals that account for discrepancies between the actual machine data under normal conditions and the baseline nominal model predictions. The hybrid model is validated against 20 Mazak milling machine experimental tests and one Haas run-to-failure experiment. The proposed anomaly detector is applied to synthetic data from simulations of the physics-based model at different operating conditions, measurement noise levels, and tool wear levels, and the methods were able to achieve an overall 92% accuracy in data with 1% noise. The anomaly detection methods based on hybrid model reduced the false alarms of either the data-driven or physical-based models alone, and are found to be capable of good online detection of tool wear.
A combination of destructive and nondestructive testing methods was utilized to evaluate the impact velocity and energy conditions that caused fracture in alumina structural ceramics. Incremental damage was produced in aluminum-backed Al{sub 2}O{sub 3} tiles using drop tower testing for low velocity impact with a high mass indenter. The majority of damage in the fractured samples consisted of radial and cone cracking. The samples were nondestructively evaluated using digital radiography, x-ray computed tomography, and ultrasound C-scan imaging. Bulk damage detected by these techniques was compared to surface damage observed by visual inspection. Qualitative evaluation of surface and bulk cracks and quantitative percent damage assessment were used to compare the integrity of the nondestructive techniques. While all three methods were able to detect surface cracks, C-scan imaging was more effective at detecting internal damage in the alumina samples and x-ray computed tomography was more effective at producing three-dimensional images of the crack patterns.
The 40Ar/39Ar dating method is among the most versatile of geochronometers, having the potential to date a broad variety of K-bearing materials spanning from the time of Earth’s formation into the historical realm. Measurements using modern noble-gas mass spectrometers are now producing 40Ar/39Ar dates with analytical uncertainties of ~1%, thereby providing precise time constraints for a wide range of geologic and extraterrestrial processes. Analyses of increasingly smaller subsamples have revealed age dispersion in many materials, including some minerals used as neutron fluence monitors. Accordingly, interpretive strategies are evolving to address observed dispersion in dates from a single sample. Moreover, inferring a geologically meaningful “age” from a measured “date” or set of dates is dependent on the geological problem being addressed and the salient assumptions associated with each set of data. We highlight requirements for collateral information that will better constrain the interpretation of 40Ar/39Ar data sets, including those associated with single-crystal fusion analyses, incremental heating experiments, and in situ analyses of microsampled domains. To ensure the utility and viability of published results, we emphasize previous recommendations for reporting 40Ar/39Ar data and the related essential metadata, with the amendment that data conform to evolving standards of being findable, accessible, interoperable, and reusable (FAIR) by both humans and computers. Our examples provide guidance for the presentation and interpretation of 40 Ar/ 39 Ar dates to maximize their interdisciplinary usage, reproducibility, and longevity.
As the core count in high-performance computing systems keeps increasing, faults are becoming common place. Check pointing addresses such faults but captures full process images even though only a subset of the process image changes between checkpoints. We have designed a hybrid check pointing technique for MPI tasks of high-performance applications. This technique alternates between full and incremental checkpoints: At incremental checkpoints, only data changed since the last checkpoint is captured. Our implementation integrates new BLCR and LAM/MPI features that complement traditional full checkpoints. This results in significantly reduced checkpoint sizes and overheads with only moderate increases in restart overhead. After accounting for cost and savings, benefits due to incremental checkpoints are an order of magnitude larger than overheads on restarts. We further derive qualitative results indicating an optimal balance between full/incremental checkpoints of our novel approach at a ratio of 1:9, which outperforms both always-full and always-incremental check pointing.
Carnegie Mellon University (CMU) under contract from Department of Energy/National Energy Technology Laboratory (DoE/NETL) and co-funding from the Northeast Gas Association (NGA), has completed the overall system design, field-trial and Magnetic Flux Leakage (MFL) sensor evaluation program for the next-generation Explorer-II (X-II) live gas main Non-destructive Evaluation (NDE) and visual inspection robot platform. The design is based on the Explorer-I prototype which was built and field-tested under a prior (also DoE- and NGA co-funded) program, and served as the validation that self-powered robots under wireless control could access and navigate live natural gas distribution mains. The X-II system design ({approx}8 ft. and 66 lbs.) was heavily based on the X-I design, yet was substantially expanded to allow the addition of NDE sensor systems (while retaining its visual inspection capability), making it a modular system, and expanding its ability to operate at pressures up to 750 psig (high-pressure and unpiggable steel-pipe distribution mains). A new electronics architecture and on-board software kernel were added to again improve system performance. A locating sonde system was integrated to allow for absolute position-referencing during inspection (coupled with external differential GPS) and emergency-locating. The power system was upgraded to utilize lithium-based battery-cells for an increase in mission-time. The resulting robot-train system with CAD renderings of the individual modules. The system architecture now relies on a dual set of end camera-modules to house the 32-bit processors (Single-Board Computer or SBC) as well as the imaging and wireless (off-board) and CAN-based (on-board) communication hardware and software systems (as well as the sonde-coil and -electronics). The drive-module (2 ea.) are still responsible for bracing (and centering) to drive in push/pull fashion the robot train into and through the pipes and obstacles. The steering modules and their arrangement, still allow the robot to configure itself to perform any-angle (up to 90 deg) turns in any orientation (incl. vertical), and enable the live launching and recovery of the system using custom fittings and a (to be developed) launch-chamber/-tube. The battery modules are used to power the system, by providing power to the robot's bus. The support modules perform the functions of centration for the rest of the train as well as odometry pickups using incremental encoding schemes. The electronics architecture is based on a distributed (8-bit) microprocessor architecture (at least 1 in ea. module) communicating to a (one of two) 32-bit SBC, which manages all video-processing, posture and motion control as well as CAN and wireless communications. The operator controls the entire system from an off-board (laptop) controller, which is in constant wireless communication with the robot train in the pipe. The sensor modules collect data and forward it to the robot operator computer (via the CAN-wireless communications chain), who then transfers it to a dedicated NDE data-storage and post-processing computer for further (real-time or off-line) analysis. The prototype robot system was built and tested indoors and outdoors, outfitted with a Remote-Field Eddy Current (RFEC) sensor integrated as its main NDE sensor modality. An angled launcher, allowing for live launching and retrieval, was also built to suit custom angled launch-fittings from TDW. The prototype vehicle and launcher systems are shown. The complete system, including the in-pipe robot train, launcher, integrated NDE-sensor and real-time video and control console and NDE-data collection and -processing and real-time display, were demonstrated to all sponsors prior to proceeding into final field-trials--the individual components and setting for said acceptance demonstration are shown. The launcher-tube was also used to verify that the vehicle system is capable of operating in high-pressure environments, and is safely deployable using proper evacuating/purging techniques for operation in the potentially explosive natural gas environment. The test-setting and environment for safety-certification of the X-II robot platform and the launch and recovery procedures, is shown. Field-trials were successfully carried out in a live steel pipeline in Northwestern Pennsylvania. The robot was launched and recovered multiple times, travelling thousands of feet and communicating in real time with video and command-and-control (C&C) data under remote operator control from a laptop, with NDE sensor-data streaming to a second computer for storage, display and post-processing. Representative images of the activities and systems used in the week-long field-trial are shown. CMU also evaluated the ability of the X-II design to be able to integrate an MFL sensor, by adding additional drive-/battery-/steering- and support-modules to extend the X-II train.