Angela A. Sodemann

A Review of Anomaly Detection in Automated Surveillance

As surveillance becomes ubiquitous, the amount of data to be processed grows along with the demand for manpower to interpret the data. A key goal of surveillance is to detect behaviors that can be considered anomalous. As a result, an extensive body of research in automated surveillance has been developed, often with the goal of automatic detection of anomalies. Research into anomaly detection in automated surveillance covers a wide range of domains, employing a vast array of techniques. This review presents an overview of recent research approaches on the topic of anomaly detection in automated surveillance. The reviewed studies are analyzed across five aspects: surveillance target, anomaly definitions and assumptions, types of sensors used and the feature extraction processes, learning methods, and modeling algorithms.

A Survey of Distance and Similarity Measures Used Within Network Intrusion Anomaly Detection

Anomaly detection (AD) use within the network intrusion detection field of research, or network intrusion AD (NIAD), is dependent on the proper use of similarity and distance measures, but the measures used are often not documented in published research. As a result, while the body of NIAD research has grown extensively, knowledge of the utility of similarity and distance measures within the field has not grown correspondingly. NIAD research covers a myriad of domains and employs a diverse array of techniques from simple k-means clustering through advanced multiagent distributed AD systems. This review presents an overview of the use of similarity and distance measures within NIAD research. The analysis provides a theoretical background in distance measures and a discussion of various types of distance measures and their uses. Exemplary uses of distance measures in published research are presented, as is the overall state of the distance measure rigor in the field. Finally, areas that require further focus on improving the distance measure rigor in the NIAD field are presented.

Intelligent Tool-Path Segmentation for Improved Stability and Reduced Machining Time in Micromilling

This paper presents the method of variable-feedrate intelligent segmentation as an enhanced approach to feedrate optimization in micromilling that overcomes detrimental scale effects in the process and leads to improved stability and decreased machining time. Due to the high tool-size to feature-size ratio present in micromilling, the maximum allowable feedrate is limited by the sampling rate of the real-time trajectory generation and motion control system. The variable-feedrate intelligent segmentation method is proposed to compensate for the feedrate limitation by intelligent selection of the interpolation technique applied to segments along the tool path in order to reduce the trajectory generation computation time and enable increased sampling frequency. The increased sampling frequency allows higher maximum feedrates providing for increased productivity and improved process stability. The performance of the novel intelligent segmentation approach was benchmarked against recent non-rational B-splines (NURBS) feedrate optimization techniques. Results from the numerical evaluation of the intelligent segmentation technique have demonstrated significant reductions in machining time, with a maximum reduction of over 50% recorded. Furthermore, the results from the study demonstrate the advantages of the intelligent segmentation method in enhancing process stability and maintaining, or marginally decreasing, process error. The variable feedrate intelligent segmentation method developed in this study provides, therefore, an enhanced methodology for path planning in high-speed, high-precision micromilling operations.

Data-driven surge map modeling for centrifugal air compressors

For the compressor operation, surge is a detrimental phenomenon of large oscillating pressure and flow in air compressors, which occurs usually at too low flow rate for a given discharge pressure. For many industrial air compressors, surge map is widely used for surge avoidance control. Field operation of centrifugal air compressors for manufacturing plant has shown that the surge map of a compressor may vary dramatically with ambient and operational conditions. In our previous work, data-driven surge map modeling method has been developed to obtain surge maps under different ambient air conditions, where the asymmetric support vector machines (ASVM) were developed for obtaining the surge map models based on actual surge test data. Support vector machine algorithms are generally computationally intensive, which may increase the complexity and cost of implementation. In this paper, a method of effectively selecting the support vectors is applied to the ASVM based surge map modeling framework. The modeling results correctly predict all gathered surge conditions with much less support vectors and lower orders kernel. About four times reduction of model complexity was resulted.

New Command Shaping Methods for Reduced Vibration of a Suspended Payload With Constrained Trolley Motion

In manufacturing environments, a common task is to quickly move a suspended payload point-to-point along a fixed overhead conveyor track without inducing significant payload vibration. Recent research in command shaping has shown remarkably effective ways to reduce the swing of a suspended payload providing the motion of the trolley is not constrained. However, the development of a command shaper where the trajectory of the trolley is constrained to follow a fixed curvilinear path has not been explored. This paper will present the development of a simple feedforward command shaper for fast, low vibration, point-to-point movement of a payload suspended from a trolley constrained to follow a fixed generalized path. The command shaping method involves modifying the command signal by convolving it with a series of impulses. Prior work has suggested command shaping to be very effective for fast, low-vibration movement of flexible systems. In this paper, command shaping methods are applied to an overhead conveyor system constrained to move along a fixed curvilinear path. Two new command shapers are presented for canceling payload vibration induced by motion of the trolley along the path. The designed Tangential Vibration (TV) shaper reduces payload vibrations induced by tangential accelerations of the trolley along the path, while the Centripetal-Tangential Vibration (CTV) shaper reduces vibrations induced by both tangential and centripetal accelerations. A key result of this study is that a command shaper having at least three impulses is required to yield zero residual vibration for motion along a curvilinear path. A simple pendulum payload attached to an actual small-scale overhead trolley following a constrained path is used to evaluate the performance of the designed command shapers. It is shown that the designed shapers significantly reduce payload swing compared to unshaped performance. An experimental sensitivity analysis shows the designed shapers are robust to system modeling errors and variations in path parameters.Copyright

Investigation of Optimal Parameter Space for High-Speed, High-Precision Micromilling

In this paper, it is shown that due to scale effects in micromilling, tool size is a key parameter to consider in parameter optimization for productivity in a high-speed, high-precision micromilling operation. In this preliminary study, a method is proposed to determine the optimal tool sizes for roughing and finishing cuts in a micromilling operation, and corresponding feedrate and spindle speed profiles to maximize productivity. An objective function is developed for minimization, subject to a set of constraints. This paper will develop the framework of the optimization method and will consider constraints for sufficient reduction of geometric errors and achievability under spindle speed power limitations in particular. An algorithm is presented to solve the objective function assuming use of a roughing tool and a finishing tool. A numerical case study is presented to illustrate the implementation of the method.Copyright

Vibration analysis for the development of resonant microbeam high-resolution vibrotactile haptic display

One type of assistive device for the blind has attempted to convert visual information into information that can be perceived through another sense, such as touch or hearing. A vibrotactile haptic display assistive device consists of an array of vibrating elements placed against the skin, allowing the blind individual to receive visual information through touch. However, these approaches have two significant technical challenges: large vibration element size and the number of microcontroller pins required for vibration control, both causing excessively low resolution of the device. Here, we propose and investigate a type of high-resolution vibrotactile haptic display which overcomes these challenges by utilizing a ‘microbeam’ as the vibrating element. These microbeams can then be actuated using only one microcontroller pin connected to a speaker or surface transducer. We propose that this approach could solve the low-resolution problem currently present in all haptic displays. In this paper, we present the results of an investigation into the manufacturability of such a device, simulation of the vibrational characteristics, and prototyping and experimental validation of the device concept.

Comparison of Cartesian and Polar Kinematic Arrangements for Compensation of Scale Effects in Micromilling

Many fields of active research such as biomedical engineering, electronics, and optics have need of small metallic parts less than 1mm in size, with features measured in hundreds or tens of microns, with tolerances as small as 0.1 micron. Such parts include devices for studying the processes in the human body, devices that can be implanted in the human body, small lenses, and other small components. Micromilling is a microscale manufacturing process that can be used to produce a wide range of small parts, including those that have complex 3-dimensional contours. Micromilling is a process that is, on the surface, similar to conventional-scale milling, except for the use of tools that are around two orders of magnitude smaller than conventional endmills, and spindle speeds that are one or two orders of magnitude faster than conventional milling spindles. However, the underlying physical processes which occur in micromilling are unique due to scale effects, which occur due to the unequal scaling of physical properties between the conventional and the micro scale. One of the more recently-uncovered scale effects in micromilling is the increased ratio of tool size to feature size [1]. This scale effect causes an exacerbation of a kind of geometric error known as chord error and places a fundamental limitation on achievable feedrates within allowable machining error constraints. In this research, we hypothesize that the increase of chord error in microscale milling can be alleviated by intelligent modification of the kinematic arrangement of the micromilling machine. Currently, all 3-axis micromilling machines are constructed with a Cartesian kinematic arrangement, in which three linear axes are mounted perpendicularly. In this paper, we propose an alternate kinematic arrangement consisting of two linear axes and one rotary axis, creating a Polar kinematic arrangement. Through numerical simulation, we show that there are distinct classes of curvilinear geometries in which the Polar kinematic arrangement is preferable, and allows significant gains in allowable feedrates and reduction in chord error, while other curvilinear geometries show reduced chord error with the Cartesian arrangement.Copyright

An Investigation Into Fixture Error Compensation in Micromilling Using Tool-Based Conductive Touch-Off

In micro-milling, decreased tool size leads to a need for tighter tolerances for fixture error in order to avoid excessive tool load and maintain machining accuracy. In 4-axis machining on a curved surface, fixture errors propagate cumulatively leading to a significant error at the tool tip. As a result a compensation approach is essential to successful microfeature production on curved surfaces. Tool stresses are shown to be highly dependent on the amount of fixture error. The scaling down of tool sizes is shown to result in an exponential increase in tool stresses. This paper proposes the use of a conductive touch-off method that utilizes the milling tool in its spindle to perform an in-situ registration mapping of positional errors. The fixturing errors are characterized using the Denavit-Hartenberg robotic linkage convention. A forward kinematic solution uses homogeneous transformation matrices to investigate the effects of fixturing errors on milling tool path errors in 4-axis micro-milling on curved surfaces. The touch-off registration measures the positional error in the tool axis direction allowing for axial tool position compensation. This results in decreased tool stresses and increased channel depth accuracy which is necessary for successful milling. A preliminary implementation of the conductive touch-off registration approach has demonstrated the efficacy of the technique when applied to production of micro-features on concave surfaces.Copyright