This PDF file contains the front matter associated with SPIE Proceedings Volume 7833, including the Title Page, Copyright information, Table of Contents, and the Conference Committee listing.

Automatic Identification System (AIS) is an unattended vessel reporting system developed for collision avoidance. Shipboard AIS equipment automatically broadcasts vessel positional data at regular intervals. The real-time position and identity data from a vessel is received by other vessels in the area thereby assisting with local navigation. As well, AIS broadcasts are beneficial to those concerned with coastal and harbour security. Land-based AIS receiving stations can also collect the AIS broadcasts. However, reception at the land station is dependent upon the ship's position relative to the receiving station. For AIS to be used as a trusted surveillance system, the characteristics of the AIS coverage area in the vicinity of the station (or stations) should be understood. This paper presents some results of a method being investigated at DRDC Atlantic (Canada) to map the AIS coverage characteristics of a dynamic AIS reception network. The method is shown to clearly distinguish AIS reception edges from those edges caused by vessel traffic patterns. The method can also be used to identify temporal changes in the coverage area, an important characteristic for local maritime security surveillance activities. Future research using the coverage estimate technique is also proposed to support surveillance activities.

There are very good automatic detection algorithms available to be used in an Automatic Target Recognition applications. However they need lots of data for training the detector for the specific use, e.g., performing an inventory of wild animals. Ongoing work use thermally correct infrared models of animals for training the detector because collecting real images from these wild animals is too expensive if even possible. This paper describes the process of designing a good IR model of the animals, and the validation process of the thermal model. Several animals are modeled using RadThermIR to be used for training detection algorithms. Animal models are based on commercially available CAD models and are initiated by temperature values from real IR measurements in several different weather conditions. The modeling extends the available set of training images by introducing different weather conditions and different poses of the animal. Fat and fur thickness of the animal is modeled with respect to climate and weather.

In this work we focus on the task to localize and track multiple non-cooperative targets by a passive antenna array and an optical sensor. Both sensor systems are mounted on a UAV and obtain bearing measurements from the targets, where the number of targets is unknown. To solve the localization and tracking problem, the imprecise but unique bearing data collected from the antenna array has to be correlated with the precise but ambiguous bearing data gained from the optical system. We perform this by a Monte Carlo realization of a multi-sensor probability hypothesis density (PHD) filter.

We describe the fabrication of a turbocharged, microelectromechanical system (MEMS) turbine. The turbine will be part of a standalone power unit and includes extra layers to connect the turbine to a generator. The project goal is to demonstrate the successful combination of several features, namely: silicon fusion bonding (SFB), a micro turbocharger [2], two rotors, mechanical alignment between two wafers [1], and the use of only one 5" silicon wafer. The dimension of the actual turbine casing will be 14mm. The turbine rotor will have a diameter of 8mm. Given these dimensions, MEMS processes are an adequate way to fabricate the device, but it will be necessary to stack up seven different layers to build the turbine, as it is not possible to construct it out of one thick wafer. SFB will be used for bonding because it permits the great precision necessary for high quality alignment. Yet a more precise alignment will be necessary between the layers that contain the turbine rotor, to decrease imbalance and guarantee operation at a very high rpm. To achieve these tight tolerances, a mechanical alignment feature announced by Liudi Jiang [1] is used. The alignment accuracy is expected to be around 200nm. Despite the fact that the turbine consists of multiple layers, it will be fabricated on only one silicon-on-insulator (SOI) wafer. As a result, all layers are exposed to the same process flow. The fabrication process includes MEMS technology as photolithography, nine deep reactive ion etching (DRIE) steps, and six SFB operations. A total of 14 masks are necessary for the fabrication.

A common miniature environment-insensitive navigation module which not only can provide the accurate position through different environment but also can easily be configured to adapt to the different type of unmanned vehicles is proposed in this paper. The module prototype is composed of a integrated MEMS inertial navigation unit using MEMS accelerometers and gyroscopes, a hard aluminum module structure with rubber isolator, a series of sensor interface of magnetometer, embedded GPS receiver, infrared sensor, vision camera, radio frequency communication etc and a FPGA based central control and navigation calculation circuit. The fabricated MEMS accelerometers and gyroscopes can resist high-g shock and have temperature drift compensation. The structure of the module uses hard aluminum with finite element analysis to find the appropriate position for sensors. All sensors without antenna are packaged in the structure with moisture, heat isolation and rubber isolator. The navigation computation scheme use the status of connected sensors to choose appropriate algorithm function to compute the navigation output. FPGA is used to be the main control and process unit of this module. Algorithms are embedded in the FPGA using the DSP core. The multiply interfaces to other sensor are implemented using the flexible configuration of the FPGA and peripheral. The conclusions are reached at last.

This paper presents a new linear inertial acceleration switch which senses inertial acceleration and gives a signal of switchpoint. It is an entire mechanical device has two particular characters: a simple structure and an environmental interference-free capability. The structure and work principle of the switch is introduced, then the design process to the spring is analyzed and simulated, and finally the rationality of this acceleration switch's design is given according to the sample's testing data. In this acceleration switch, the elastic component is a leaf spring, and the mass component is a standard steel ball. The spring and the ball are separated instead of rigidly connected, which make the whole structure is simple. When the switch is on the work direction the ball and the spring are interact, and the spring is on work; when the switch isn't on the work direction, the ball and the spring are separated; environmental external force is on the mass instead of on the spring. The spring is insusceptible on this condition. This particularity determines that the switch is highly environmental interference-free, and doesn't easily affected by environmental influence. Some parameters of the inertial switch are given as followings: (1) Overall dimension of the inertial switch is about 28mm×12mm×12mm; (2) systemic precision of the inertial switch is 1.5%; (3) the inertial switch can endure 0.2g²/Hz stochastic vibration. It is suggested that this inertial switch can be applied in high consequence system.

A novel range detection technique combines optical pulse modulation patterns with signal cross-correlation to produce an accurate range estimate from low power signals. The cross-correlation peak is analyzed by a post-processing algorithm such that the phase delay is proportional to the range to target. This technique produces a stable range estimate from noisy signals. The advantage is higher accuracy obtained with relatively low optical power transmitted. The technique is useful for low cost, low power and low mass sensors suitable for tactical use. The signal coding technique allows applications including IFF and battlefield identification systems.

This paper describes a prototype demonstration of a high bandwidth data link between the fuselage of an aircraft and a helmet mounted display. A single data receiver, powered by battery and equipped with a light-collecting optical antenna to increase optical gain, is worn on the body of the pilot, with a fast-modulated laser transmitter mounted in the pilot's seat area. The combination covered the expected range of body movement that a pilot typically undergoes during a flight. Uncompressed, ~140Mbps video data is streamed over the free-space link to a BAE Systems helmet mounted display (Q-Sight™) worn by the pilot.

We describe the development of flexi-circuit addressed, high resolution small aperture Bimorph Deformable Mirror (BDM) suitable for a satellite to ground optical communications system. In the first phase, the base of an existing BDM was modified to enable integration onto a commercially available tip-tilt stage. Evaluation demonstrated that there was no reduction in performance as a result of the integration. In the current phase, a 100 element BDM with a 10mm optical aperture suitable for integration onto a bespoke tip-tilt platform has been designed and assembled. The results of initial tests are presented.

This paper presents an overview of a human-robotic system under development at Cornell which is capable of mapping an unknown environment, as well as discovering, tracking, and neutralizing several static and dynamic objects of interest. In addition, the robots can coordinate their individual tasks with one another without overly burdening a human operator. The testbed utilizes the Segway RMP platform, with lidar, vision, IMU and GPS sensors. The software draws from autonomous systems research, specifically in the areas of pose estimation, target detection and tracking, motion and behavioral planning, and human robot interaction. This paper also details experimental scenarios of mapping, tracking, and neutralization presented by way of pictures, data, and movies.

Dead Reckoning (DR) is the process of estimating a robot's current position based upon a previously determined position, and advancing that position based upon known speed and direction over time. It is therefore a simple way for an autonomous mobile robot to navigation within a known environment such as a building where measurements have been taken and a predetermined route planned based upon which doors (or areas) the robot would have enough force to enter. Discussed here is the design of a DR navigation system in Altera's DSP Builder graphical design process. The wheel circumference to the step size of stepper motor used to drive the robot are related and so this ratio can be easily changed to easily accommodate changes to the physical design of a robot with minimal changes to the software. The robot calculates its position in relation to the DR map by means of the number of revolutions of the wheels via odometry, in this situation there is no assumed wheel slippage that would induce an accumulative error in the system overtime. The navigation works by using a series of counters, each corresponding to a measurement taken from the environment, and are controlled by a master counter to trigger the correct counter at the appropriate time given the position of robot in the DR map. Each counter has extra safeguards built into them on their enables and outputs to ensure they only count at the correct time and to avoid clashes within the system. The accuracy of the navigation is discussed after the virtual route is plotted in MATLAB as a visual record in addition to how feedback loops, identification of known objects (such as fire safety doors that it would navigate through), and visual object avoidance could later be added to augment the system. The advantages of such a system are that it has the potential to upload different DR maps so that the end robot for can be used in new environments easily.

This paper explores representations for capturing the anticipation of other objects by an autonomous robot in an urban environment. Predictive Gaussian mixture models are proposed due to their ability to probabilistically capture continuous and discrete obstacle behavior; the predictive system uses the probabilistic output of a tracking system (current obstacle location), and map (with lanes and intersections). The probabilistic tracking and anticipated motion are integrated into an optimized path planner. This paper explores various levels of model abstraction to understand how complex these predictive models must be in order to create a more robust path planning algorithm.

Unattended ground sensors (UGS) provide the capability to inexpensively secure remote borders and other areas of interest. However, the presence of normal animal activity can often trigger a false alarm. Accurately detecting humans and distinguishing them from natural fauna is an important issue in security applications to reduce false alarm rates and improve the probability of detection. In particular, it is important to detect and classify people who are moving in remote locations and transmit back detections and analysis over extended periods at a low cost and with minimal maintenance. We developed and demonstrate a compact radar technology that is scalable to a variety of ultra-lightweight and low-power platforms for wide area persistent surveillance as an unattended, unmanned, and man-portable ground sensor. The radar uses micro-Doppler processing to characterize the tracks of moving targets and to then eliminate unimportant detections due to animals as well as characterize the activity of human detections. False alarms from sensors are a major liability that hinders widespread use. Incorporating rudimentary intelligence into sensors can reduce false alarms but can also result in a reduced probability of detection. Allowing an initial classification that can be updated with new observations and tracked over time provides a more robust framework for false alarm reduction at the cost of additional sensor observations. This paper explores these tradeoffs with a small radar sensor for border security. Multiple measurements were done to try to characterize the micro-Doppler of human versus animal and vehicular motion across a range of activities. Measurements were taken at the multiple sites with realistic but low levels of clutter. Animals move with a quadrupedal motion, which can be distinguished from the bipedal human motion. The micro-Doppler of a vehicle with rotating parts is also shown, along with ground truth images. Comparisons show large variations for different types of motion by the same type of animal. This paper presents the system and data on humans, vehicles, and animals at multiple angles and directions of motion, demonstrates the signal processing approach that makes the targets visually recognizable, verifies that the UGS radar has enough micro-Doppler capability to distinguish between humans, vehicles, and animals, and analyzes the probability of correct classification.

Stereo vision is a situation where an imaging system has two or more cameras in order to make it more robust by mimicking the human vision system. By using two inputs, knowledge of their own relative geometry can be exploited to derive depth information from the two views they receive. 3D co-ordinates of an object in an observed scene can be computed from the intersection of the two sets of rays. Presented here is the development of a stereo vision system to focus on an object at the centre of a baseline between two cameras at varying distances. This has been developed primarily for use on a Field Programmable Gate Array (FPGA) but an adaptation of this developed methodology is also presented for use with a PUMA 560 Robotic Manipulator with a single camera attachment. The two main vision systems considered here are a fixed baseline with an object moving at varying distances from this baseline, and a system with a fixed distance and a varying baseline. These two differing situations provide enough data so that the co-efficient variables that determine the system operation can be calibrated automatically with only the baseline value needing to be entered, the system performs all the required calculations for the user for use with a baseline of any distance. The limits of system with regards to the focusing accuracy obtained are also presented along with how the PUMA 560 controls its joints for the stereo vision and how it moves from one position to another to attend stereo vision compared to the two camera system for the FPGA. The benefits of such a system for range finding in mobile robotics are discussed and how this approach is more advantageous when compared against laser range finders or echolocation using ultrasonics.

The paper presents the concept of optoelectronic devices for human protection in rescue activity. The system consists of an ground robots with predicted sensor. The multisensor construction of the system ensures significant improvement of security of using on-situ like chemical or explosive sensors. The article show a various scenario of use for individual sensor in system PROTEUS.

State-of-the-art Inertial Navigation Systems (INS) based on Micro-Electro-Mechanical Systems (MEMS) have a lack of precision especially in GPS denied environments like urban canyons or in pure indoor missions. The proposed Optical Navigation System (ONS) provides bias free ego-motion estimates using triple redundant sensor information. In combination with a model based state prediction our system is able to estimate velocity, position and attitude of an arbitrary aircraft. Simulating a high performance flow-field estimator the algorithm can compete with conventional low-cost INS. By using measured velocities instead of accelerations the system states drift behavior is not as distinctive as for an INS.

To allow Unmanned Aircraft Systems (UAS) accessing National Airspace System (NAS) "Equivalent levels of safety" to the ones of human vision must be guaranteed. Therefore, an appropriate "Sense and Avoid" technology must be developed that is capable of detecting, tracking, and avoiding obstacles. The Department of Aerospace Engineering at University of Naples has been involved in a project funded by the Italian Aerospace Research Centre (CIRA) for the realization of a prototypical "Obstacle Detection & Identification" (ODID) System. It is installed onboard a Very Light Aircraft (VLA) and it is characterized by a hierarchical sensor configuration in which the radar is the main sensor while EO cameras are the auxiliary ones in order to increase accuracy and data rate so that anti-collision requirements are fulfilled. This paper focuses on the Image Processing algorithm for the panchromatic camera. Among the several techniques listed in literature the edge detection - labeling one resulted as the best compromise in terms of computational load, detection range, false alarm rate, miss detection rate and adaptability at different background luminosity conditions. Moreover it has been customized in order to allow for reliable operation in a wide range of flight and luminance configurations and it has been tested and run on a sequence of real images taken during flight tests. At the end, a table that summarizes those results is presented. Indeed, the output tracking measurements accuracy increases by an order of magnitude with respect to standalone radar one.

Military fighter pilots have to make suitable decisions fast in an environment where continuously increasing flows of information from sensors, team members and databases are provided. Not only do the huge amounts of data aggravate the pilots' decision making process: time-pressure, presence of uncertain data and high workload are factors that can worsen the performance of pilot decision making. In this paper, initial ideas of how to support the pilots accomplishing their tasks are presented. Results from interviews with two fighter pilots are described as well as a discussion about how these results can guide the design of a military fighter pilot decision support system, with focus on team cooperation.

The imbalanced learning problem (learning from imbalanced data) presents a significant new challenge to the pattern recognition and machine learning society because in most instances real-world data is imbalanced. When considering military applications, the imbalanced learning problem becomes much more critical because such skewed distributions normally carry the most interesting and critical information. This critical information is necessary to support the decision-making process in battlefield scenarios, such as anomaly or intrusion detection. The fundamental issue with imbalanced learning is the ability of imbalanced data to compromise the performance of standard learning algorithms, which assume balanced class distributions or equal misclassification penalty costs. Therefore, when presented with complex imbalanced data sets these algorithms may not be able to properly represent the distributive characteristics of the data. In this paper we present an empirical study of several popular imbalanced learning algorithms on an army relevant data set. Specifically we will conduct various experiments with SMOTE (Synthetic Minority Over-Sampling Technique), ADASYN (Adaptive Synthetic Sampling), SMOTEBoost (Synthetic Minority Over-Sampling in Boosting), and AdaCost (Misclassification Cost-Sensitive Boosting method) schemes. Detailed experimental settings and simulation results are presented in this work, and a brief discussion of future research opportunities/challenges is also presented.

MWIR photon detector in the mid-infrared wavelength (2-5 μm) range is developed using crystalline silicon carbide substrates. SiC, which is a wideband gap semiconductor, is laser-doped to create a dopant energy level corresponding to a quantum of energy for the required operating wavelength of the detector. The photons of the objects in the field of view excite the electrons of the detector, leading to changes in the refractive index. This change in the optical property of the detector can be measured remotely with a laser beam, such as a He-Ne laser beam of wavelength 632.8 nm, which makes it a wireless detector. While many IR detectors require cryogenic cooling (77 K) to suppress thermal generationrecombination processes in order to operate with good detectivity, the SiC-based detector can operate at room temperature with excellent performance. An n-type 4H-SiC substrate has been doped with Ga by a laser doping technique to create a detector element for the MWIR wavelength of 4.21 μm corresponding to the photon energy 0.30 eV. The dopant energy level is confirmed by optical absorption measurements. The change in the refractive index is studied as a function of absorbed irradiance on the detector. The experimental result shows that the Ga-doped 4H-SiC sample can be used for MWIR detectors.

Morphological Scene Change Detection (MSCD) is a process typically tasked at detecting relevant changes in a guarded environment for security applications. This can be implemented on a Field Programmable Gate Array (FPGA) by a combination of binary differences based around exclusive-OR (XOR) gates, mathematical morphology and a crucial threshold setting. The additional ability to set up the system in virtually any location due to the FPGA makes it ideal for insertion into an autonomous mobile robot for patrol duties. However, security is not the only potential of this robust algorithm. This paper details how such a system can be used for the detection of leaks in piping for use in the process and chemical industries and could be deployed as stated in the above manner. The test substance in this work was water, which was pumped either as a liquid or as low pressure steam through a simple pipe configuration with holes at set points to simulate the leaks. These holes were situated randomly at either the center of a pipe (in order to simulate an impact to it) or at a joint or corner (to simulate a failed weld). Imagery of the resultant leaks, which were visualised as drips or the accumulation of steam, which where analysed using MATLAB to determine their pixel volume in order to calibrate the trigger for the MSCD. The triggering mechanism is adaptive to make it possible in theory for the type of leak to be determined by the number of pixels in the threshold of the image and a numerical output signal to state which of the leak situations is being observed. The system was designed using the DSP Builder package from Altera so that its graphical nature is easily comprehensible to the non-embedded system designer. Furthermore, all the data from the DSP Builder simulation underwent verification against MATLAB comparisons using the image processing toolbox in order to validate the results.

Fiber optical interferometers belong to highly sensitive equipments that are able to measure slight changes like distortion of shape, temperature and electric field variation and etc. Their great advantage is that they are insensitive on ageing component, from which they are composed of. It is in virtue of herewith, that there are evaluated no changes in optical signal intensity but number interference fringes. To monitor the movement of persons, eventually to analyze the changes in state of motion we developed method based on analysis the dynamic changes in interferometric pattern. We have used Mach- Zehnder interferometer with conventional SM fibers excited with the DFB laser at wavelength of 1550 nm. It was terminated with optical receiver containing InGaAs PIN photodiode. Its output was brought into measuring card module that performs on FFT of the received interferometer signal. The signal rises with the composition of two waves passing through single interferometer arm. The optical fiber SMF 28e in one arm is referential; the second one is positioned on measuring slab at dimensions of 1x2m. A movement of persons over the slab was monitored, signal processed with FFT and frequency spectra were evaluated. They rose owing to dynamic changes of interferometric pattern. The results reflect that the individual subjects passing through slab embody characteristic frequency spectra, which are individual for particular persons. The scope of measuring frequencies proceeded from zero to 10 KHz. It was also displayed in experiments that the experimental subjects, who walked around the slab and at the same time they have had changed their state of motion (knee joint fixation), embodied characteristic changes in their frequency spectra. At experiments the stability of interferometric patterns was evaluated as from time aspects, so from the view of repeated identical experiments. Two kinds of balls (tennis and ping-pong) were used to plot the repeatability measurements and the gained spectra at repeated drops of balls were compared. Those stroked upon the same place and from the same elevation and dispersion of the obtained frequency spectra was evaluated. These experiments were performed on the series of 20 repeated drops from highs of 0,5 and 1m. The evaluation of experiments displayed that the dispersion of measured values is lower than 4%.

Effective situation awareness is a critical element for decision support in a wide range of military and para-military operational surveillance scenarios. Effective situation awareness in a surveillance scenario can greatly increase operational effectiveness, by improving the quality and timeliness of decisions. In this paper we outline a three level integrated design approach to promote situation awareness. Our approach allows deployed wireless sensor nodes to efficiently self-organise into dynamic clusters, based on a current common perceived threat situation (context). Firstly our distributed predator aware situation assessment system (PORTENT) models, detects and presents, in terms of quality of information (QoI), potential situations occurring within an uncertain environment. Secondly, we utilise a Bayesian belief network to understand the significance associated with the potential situation. Finally in order to obtain a better shared awareness we have developed a "context aware" service protocol that supports group formation and efficient management of sensor network assets. By combining this three level approach, we present our VIGILANT "situation aware" QoI interest group system. Extensive simulations have been undertaken to verify the VIGILANT concept, to demonstrate the effectiveness of our approach, in improving performance for network management efficiency, through utilisation of a shared "context" service provision time and QoI surveillance presentation.

Free-space optics (FSO) holds the potential for high bandwidth communication, but atmospheric conditions can significantly affect the capability of this type of communication system to transfer information consistently and operate effectively. The effects of atmosphere on FSO communication and consequent optimal wavelength range for transmission are investigated through MODTRAN-based modeling of 1.55 μm transmission for multiple elevation angles in atmospheric weather conditions including clear maritime, desert extinction, and various levels of rain and fog, to simulate surface-to-surface and surface-to-air FSO communication networks. Furthermore, atmospheric, free-space, and scintillation losses are analyzed for optical path lengths of 2 km to determine minimum transmit power required for successful data reception. In addition, FSO transmitter and receiver circuits were designed to optically relay analog video signals and their operation verified at path distances of up to 130 m. Using advanced laser sources to provide illumination at infrared wavelengths, particularly around the eye-safe 1.55 μm wavelength, it should be possible to overcome many transmission limitations associated with atmospheric conditions such as adverse weather and turbulence.