Anthony J. Pinar

Low-Cost Open-Source Voltage and Current Monitor for Gas Metal Arc Weld 3D Printing

Arduino open-source microcontrollers are well known in sensor applications for scientific equipment and for controlling RepRap 3D printers. Recently low-cost open-source gas metal arc weld (GMAW) RepRap 3D printers have been developed. The entry-level welders used have minimal controls and therefore lack any real-time measurement of welder voltage or current. The preliminary work on process optimization of GMAW 3D printers requires a low-cost sensor and data logger system to measure welder current and voltage. This paper reports on the development of a low-cost open-source power measurement sensor system based on Arduino architecture. The sensor system was designed, built, and tested with two entry-level MIG welders. The full bill of materials and open source designs are provided. Voltage and current were measured while making stepwise adjustments to the manual voltage setting on the welder. Three conditions were tested while welding with steel and aluminum wire on steel substrates to assess the role of electrode material, shield gas, and welding velocity. The results showed that the open source sensor circuit performed as designed and could be constructed for <

Feature and decision level fusion using multiple kernel learning and fuzzy integrals

100 in components representing a significant potential value through lateral scaling and replication in the 3D printing community.

Efficient Multiple Kernel Classification Using Feature and Decision Level Fusion

Kernel methods for classification is a well-studied area in which data are implicitly mapped from a lower-dimensional space to a higher-dimensional space to improve classification accuracy. However, for most kernel methods, one must still choose a kernel to use for the problem. Since there is, in general, no way of knowing which kernel is the best, multiple kernel learning (MKL) is a technique used to learn the aggregation of a set of valid kernels into a single (ideally) superior kernel. The aggregation can be done using weighted sums of the pre-computed kernels, but determining the summation weights is not a trivial task. A popular and successful approach to this problem is MKL-group lasso (MKLGL), where the weights and classification surface are simultaneously solved by iteratively optimizing a min-max optimization until convergence. In this work, we propose an ℓp-normed genetic algorithm MKL (GAMKLp), which uses a genetic algorithm to learn the weights of a set of pre-computed kernel matrices for use with MKL classification. We prove that this approach is equivalent to a previously proposed fuzzy integral aggregation of multiple kernels called fuzzy integral: genetic algorithm (FIGA). A second algorithm, which we call decision-level fuzzy integral MKL (DeFIMKL), is also proposed, where a fuzzy measure with respect to the fuzzy Choquet integral is learned via quadratic programming, and the decision value-viz., the class label-is computed using the fuzzy Choquet integral aggregation. Experiments on several benchmark data sets show that our proposed algorithms can outperform MKLGL when applied to support vector machine (SVM)-based classification.

Deep belief networks for false alarm rejection in forward-looking ground-penetrating radar

Kernel methods for classification is a well-studied area in which data are implicitly mapped from a lower-dimensional space to a higher dimensional space to improve classification accuracy. However, for most kernel methods, one must still choose a kernel to use for the problem. Since there is, in general, no way of knowing which kernel is the best, multiple kernel learning (MKL) is a technique used to learn the aggregation of a set of valid kernels into a single (ideally) superior kernel. The aggregation can be done using weighted sums of the precomputed kernels, but determining the summation weights is not a trivial task. Furthermore, MKL does not work well with large datasets because of limited storage space and prediction speed. In this paper, we address all three of these multiple kernel challenges. First, we introduce a new linear feature level fusion technique and learning algorithm, GAMKLp. Second, we put forth three new algorithms, DeFIMKL, DeGAMKL, and DeLSMKL, for nonlinear fusion of kernels at the decision level. To address MKLs storage and speed drawbacks, we apply the Nystrom approximation to the kernel matrices. We compare our methods to a successful and state-of-the-art technique called MKL-group lasso (MKLGL), and experiments on several benchmark datasets show that some of our proposed algorithms outperform MKLGL when applied to support vector machine (SVM)-based classification. However, to no surprise, there does not seem to be a global winner but instead different strategies that a user can employ. Experiments with our kernel approximation method show that we can routinely discard most of the training data and at least double prediction speed without sacrificing classification accuracy. These results suggest that MKL-based classification techniques can be applied to big data efficiently, which is confirmed by an experiment using a large dataset.

Multi-band sensor-fused explosive hazards detection in forward-looking ground penetrating radar

Explosive hazards are one of the most deadly threats in modern conflicts. The U.S. Army is interested in a reliable way to detect these hazards at range. A promising way of accomplishing this task is using a forward-looking ground-penetrating radar (FLGPR) system. Recently, the Army has been testing a system that utilizes both L-band and X-band radar arrays on a vehicle mounted platform. Using data from this system, we sought to improve the performance of a constant false-alarm-rate (CFAR) prescreener through the use of a deep belief network (DBN). DBNs have also been shown to perform exceptionally well at generalized anomaly detection. They combine unsupervised pre-training with supervised fine-tuning to generate low-dimensional representations of high-dimensional input data. We seek to take advantage of these two properties by training a DBN on the features of the CFAR prescreener’s false alarms (FAs) and then use that DBN to separate FAs from true positives. Our analysis shows that this method improves the detection statistics significantly. By training the DBN on a combination of image features, we were able to significantly increase the probability of detection while maintaining a nominal number of false alarms per square meter. Our research shows that DBNs are a good candidate for improving detection rates in FLGPR systems.

Highly Maneuverable Low-Cost Underwater Glider: Design and Development

Explosive hazard detection and remediation is a pertinent area of interest for the U.S. Army. There are many types of detection methods that the Army has or is currently investigating, including ground-penetrating radar, thermal and visible spectrum cameras, acoustic arrays, laser vibrometers, etc. Since standoff range is an important characteristic for sensor performance, forward-looking ground-penetrating radar has been investigated for some time. Recently, the Army has begun testing a forward-looking system that combines L-band and X-band radar arrays. Our work focuses on developing imaging and detection methods for this sensor-fused system. In this paper, we investigate approaches that fuse L-band radar and X-band radar for explosive hazard detection and false alarm rejection. We use multiple kernel learning with support vector machines as the classification method and histogram of gradients (HOG) and local statistics as the main feature descriptors. We also perform preliminary testing on a context aware approach for detection. Results on government furnished data show that our false alarm rejection method improves area-under-ROC by up to 158%.

A multi-level motion controller for low-cost Underwater Gliders

This letter presents the design and potential impact of the developed Research Oriented Underwater Glider for Hands-on Investigative Engineering (ROUGHIE). The ROUGHIE is an open-source, highly maneuverable, and low-cost vehicle that enables rapid development and testing of new hardware and software. ROUGHIE is an internally actuated glider capable of performing steady sawtooth glides in shallow water down to 3 m, tight turns with a minimum radius of 3 m, and a minimum endurance of 60 h. The novelty of this study is twofold: 1) a rail-based design to facilitate modularity and ease of assembly and 2) an effective internal rotary mass mechanism to increase maneuverability and perform tight turns. The ROUGHIE design strategically uses 3D printed plastic parts in low stress situations, which allows extreme design flexibility and enables tightly packed modules that can be easily customized.

Effective Turning Motion Control of Internally Actuated Autonomous Underwater Vehicles

An underwater glider named ROUGHIE (Research Oriented Underwater Glider for Hands-on Investigative Engineering) is designed and manufactured to provide a test platform and framework for experimental underwater automation. This paper presents an efficient multi-level motion controller that can be used to enhance underwater glider control systems or easily modified for additional sensing, computing, or other requirements for advanced automation design testing. The ultimate goal is to have a fleet of modular and inexpensive test platforms for addressing the issues that currently limit the use of autonomous underwater vehicles (AUVs). Producing a low-cost vehicle with maneuvering capabilities and a straightforward expansion path will permit easy experimentation and testing of different approaches to improve underwater automation.

A novel roll mechanism to increase maneuverability of autonomous underwater vehicles in shallow water

This paper presents a novel roll mechanism and an efficient control strategy for internally actuated autonomous underwater vehicles (AUVs). The developed control algorithms are tested on Michigan Tech’s custom research glider, ROUGHIE (Research Oriented Underwater Glider for Hands-on Investigative Engineering), in a controlled environment. The ROUGHIE’s design parameters and operational constraints were driven by its requirement to be man portable, expandable, and maneuverable in shallow water. As an underwater glider, the ROUGHIE is underactuated with direct control of only depth, pitch, and roll. A switching control method is implemented on the ROUGHIE to improve its maneuverability, enabling smooth transitions between different motion patterns. This approach uses multiple feedforward-feedback controllers. Different aspects of the roll mechanism and the effectiveness of the controller on turning motion are discussed based on experimental results. The results illustrate that the ROUGHIE is capable of achieving tight turns with a radius of 2.4 meters in less than 3 meters of water, or one order of magnitude improvement on existing internally actuated platforms. The developed roll mechanism is not specific to underwater gliders and is applicable to all AUVs, especially at lower speeds and in shallower water when external rudder is less effective in maneuvering the vehicle.

Approach to explosive hazard detection using sensor fusion and multiple kernel learning with downward-looking GPR and EMI sensor data

This paper presents a novel roll mechanism and an efficient control strategy for the roll and pitch of internally actuated autonomous underwater vehicles (AUVs) including most underwater gliders (UGs). The proposed design and approach increases maneuverability which is essential for operating in shallow water or crowded harbors. The design is implemented on Michigan Techs research UG ROUGHIE (Research Oriented Underwater Glider for Hands-on Investigative Engineering) and the performance is validated. The experimental results demonstrate that ROUGHIE is capable of tight turn radii down to approximately twice the vehicle length in shallow water.