Motion Coherent Tracking Using Multi-label MRF Optimization

We present a novel off-line algorithm for target segmentation and tracking in video. In our approach, video data is represented by a multi-label Markov Random Field model, and segmentation is accomplished by finding the minimum energy label assignment. We propose a novel energy formulation which incorporates both segmentation and motion estimation in a single framework. Our energy functions enforce motion coherence both within and across frames. We utilize state-of-the-art methods to efficiently optimize over a large number of discrete labels. In addition, we introduce a new ground-truth dataset, called Georgia Tech Segmentation and Tracking Dataset (GT-SegTrack), for the evaluation of segmentation accuracy in video tracking. We compare our method with several recent on-line tracking algorithms and provide quantitative and qualitative performance comparisons.     

Inclination of mold pattern’s sidewalls by combined technique with photolithography at defocus-positions and electroforming

We succeeded in achieving inclined sidewalls in a positive-tone photoresist structure by adjusting the focus offsets in a projection ultraviolet (UV) lithography system. In our experiments, highly precise patterns on Ni-electroformed mold were replicated from a photoresist master. The practicality of the mold was then evaluated by thermal nanoimprint experiments on polycarbonate (PC). When the focus offsets became small by moving the UV image plane (defined by wafer surface) closer to the lens the inclined angles of the photoresist walls increased. In this work, the release force was measured using a desktop nanoimprint system equipped with a load cell that measured not only the compression power but also the tensile force in the de-molding process. Starting from the glass transition temperature T _g (144°C) of PC, the heating temperature in the molding process was changed by increments of 10°C resulting in T _g + 10, T _g + 20, and T _g + 30°C. The result showed that the release force decreased with increasing incline angles. It was observed that the more inclined the sidewalls of the mold pattern were, further small the release forces became, and the mold pattern became to be defended from deformation.     

Coalition structure generation in cooperative games with compact representations

This paper presents a new way of formalizing the coalition structure generation problem (CSG) so that we can apply constraint optimization techniques to it. Forming effective coalitions is a major research challenge in AI and multi-agent systems. CSG involves partitioning a set of agents into coalitions to maximize social surplus. Traditionally, the input of the CSG problem is a black-box function called a characteristic function, which takes a coalition as input and returns the value of the coalition. As a result, applying constraint optimization techniques to this problem has been infeasible. However, characteristic functions that appear in practice often can be represented concisely by a set of rules, rather than treating the function as a black box. Then we can solve the CSG problem more efficiently by directly applying constraint optimization techniques to this compact representation. We present new formalizations of the CSG problem by utilizing recently developed compact representation schemes for characteristic functions. We first characterize the complexity of CSG under these representation schemes. In this context, the complexity is driven more by the number of rules than by the number of agents. As an initial step toward developing efficient constraint optimization algorithms for solving the CSG problem, we also develop mixed integer programming formulations and show that an off-the-shelf optimization package can perform reasonably well.     

Compensation for dead band of force measurement based on the coefficient of restitution in a hybrid simulator

In order to develop space robotic technologies, the feasibility and reliability of the systems have to be verified by repetitive operation tests. A hybrid simulator (also referred to as a Hardware-in-the-loop simulator) is one of the effective ways to examine orbital operations on the ground. The simulator embeds a hardware experiment in a numerical simulation loop. However, it is known that the hybrid simulator has a serious problem that energy increase arises in contact with a hardware experiment in the loop. In general, the energy increase in the hybrid simulator occurs due to the dead time in the system. However, this paper presents that dead band in force/torque measuring also causes the energy increase. The dead band in the force/torque measuring is necessary to avoid unexpected motion of the hybrid simulator due to the noise data in the force/torque sensor. This paper proposes two compensation methods for the energy increase problem due to the dead band. The first method is to insert a virtual damper to absorb the increased energy during the contact. The second method is to extrapolate force/torque data that is lost through the dead band processing right after the contact. The experimental verification with uniaxial contact in hybrid simulator is carried out to validate the proposed compensation methods.     

Appearance-based navigation and homing for autonomous mobile robot

In this paper, we develop an algorithm for navigating a mobile robot using the visual potential. The visual potential is computed from an image sequence and optical flow computed from successive images captured by a camera mounted on the robot, that is, the visual potential for navigation is computed from appearances of the workspace observed as an image sequence. The direction to the destination is provided at the initial position of the robot. The robot dynamically selects a local pathway to the destination without collision with obstacles and without any knowledge of the robot workspace. Furthermore, the guidance algorithm to destination allows the mobile robot to return from the destination to the initial position. We present the experimental results of navigation and homing in synthetic and real environments.     

Radar remote sensing of springtime near-surface soil thaw events at mid-latitudes

Detection of springtime near-surface soil thaw events in mid-latitude areas is important for understanding the near-surface earth surface system. Radar backscatter time series, such as QuikSCAT K_u-band, are valuable for surface thaw detection in permafrost areas, but their usefulness in mid-latitudes remains unverified. In order to validate the application of radar backscatter time series for surface thaw detection in mid-latitude areas, we propose a new multi-step method based on the significant signature of radar backscatter time series when springtime soil thaw events happen. The method mainly focuses on the estimated geographical boundary of thaw events and the detection of the primary thaw date; the duration of the freeze–thaw transition period is a lesser focus. The method is applied to soil freeze–thaw conditions in a mid-latitude area – northern China and Mongolia – including humid land and dryland, and achieves a reliability of R ²?=?0.678 ± 0.021 (P < 0.01) by comparison with ground truth measurements. The results indicate that elevation/temperature and soil moisture conditions are the key drivers of the timing of springtime soil thaw events. The success of our study demonstrates the potential for our method to be applied in mid-latitude areas, and widens the application potential of radar backscatter time series beyond high north and permafrost studies.     

User-perceived reliability of unrepairable shared protection systems with functionally identical units

In this article, we investigate the reliability of M-for-N (M:N) shared protection systems. We focus on the reliability that is perceived by an end user of one of N units. We assume that any failed unit is instantly replaced by one of the M units (if available). We describe the effectiveness of such a protection system in a quantitative manner under the condition that the failed units are not repairable. Mathematical analysis gives the closed-form solution of the reliability and mean time to failure (MTTF). We also analyse several numerical examples of the reliability and MTTF. This result can be applied, for example, to the analysis and design of an integrated circuit consisting of redundant backup components. In such a device, repairing a failed component is unrealistic. The analysis provides useful information for the design for general shared protection systems in which the failed units are not repaired.     

Elucidation of fatigue characteristics and fracture mechanism of friction stir spot‐welded tension–shear joint steels

The fatigue life and fracture mechanism of friction stir spot welded tension–shear joints using 590‐MPa class steel as a base material under constant‐amplitude conditions were investigated with focus on welding dimension variations caused by tool wear. The fatigue limit of the friction stir spot welding (FSSW) joint used for this study is significantly low compared with the static tensile strength of the joint itself. It was clarified that the FSSW joint in this study exhibited two different failure morphologies regardless of the applied load level: base metal fracture and weld area fracture. Although the welding state changes due to the tool wear phenomenon that produce two types of fracture modes in relation to the welding rip diameter, they have no effect on the fatigue strength, regardless of the applied load.