Fault prediction is critically important for many mobile equipments such as vehicles, ships and spacecrafts. Sensors deployed on these equipments continuously collect the status data, which are usually multivariate time series data. It is challenging to accurately predict the failure of the equipments based on the generated multivairate time series due to the complex correlations among the variables and the dynamic operation conditions. Though many methods have been proposed, they are not effective to provide an interpretable and accurate fault prediction result. This paper proposes a two-stage Interpretable Fault Prediction method based on Anomaly Detection and Anomaly Accumulation, called IFP-ADAC. Specially, we first design an anomaly detection module based on Generative Adversarial Nets due to the lack of samples. The generator captures the correlations among multiple variables and the temporal dependency within each variable jointly. Second, we design an anomaly accumulation model based on LSTM to capture the anomaly growth pattern, and the attention mechanism has been introduced to consider the severity of the detected anomalies. Compared with the end- to-end methods, our two-stage fault prediction method based on anomaly detection and accumulation has better interpretability. Extensive experiments conducted on two real-world datasets show the superior performance of our method.

Fuel-efficient route recommendation has been increasingly valuable for both energy conservation and environmental protection. Most existing methods analyze fuel consumption factors from the short-distance trajectories. However, due to the differences in the road network structure and route composition between long-distance and short-distance trajectories, directly using these methods to recommend fuel-efficient routes for long-distance drivers is less effective. In addition, previous works usually adopt heuristic algorithms due to efficiency, but empirically set the heuristic functions, which makes it difficult to integrate various influencing factors appropriately. In this paper, we propose a novel fuel-efficient route recommendation model for long-distance driving, LDFeRR. We first identify the potential factors that affect fuel consumption over long distances based on historical trajectories. To fully exploit these factors to estimate fuel consumption and further provide reliable recommendations, we propose to integrate deep learning methods with heuristic algorithm. Specially, we use a multi-layer perceptron (MLP) to predict the fuel consumption of a single road segment. We also employ an attention-based bidirectional gated recurrent unit (Att-BiGRU) to estimate the fuel consumption between two locations connected by routes. In this way, the two cost functions of the classic heuristic algorithm, i.e., A* algorithm, are automatically learned. Extensive experiments on the large real-world dataset demonstrate the effectiveness of our proposed model.

Many rapid fabrication technologies are directed towards layer wise printing or laser based prototyping. We propose WireFab, a rapid modeling and prototyping system that uses bent metal wires as the structure framework. WireFab approximates both the skeletal articulation and the skin appearance of the corresponding virtual skin meshes, and it allows users to personalize the designs by (1) specifying joint positions and part segmentations, (2) defining joint types and motion ranges to build a wire-based skeletal model, and (3) abstracting the segmented meshes into mixed-dimensional appearance patterns or attachments.

A new visibility graph-based algorithm is presented for computing the inner distances of a 3D shape represented by a volumetric model. The inner distance is defined as the length of the shortest path between landmark points within the shape. The inner distance is robust to articulation and can reflect the deformation of a shape structure well without an explicit decomposition. Our method is based on the visibility graph approach. To check the visibility between pairwise points, we propose a novel, fast, and robust visibility checking algorithm based on a clustering technique which operates directly on the volumetric model without any surface reconstruction procedure, where an octree is used for accelerating the computation. The inner distance can be used as a replacement for other distance measures to build a more accurate description for complex shapes, especially for those with articulated parts. The binary executable program for the Windows platform is available from https://engineering.purdue.edu/PRECISE/VMID.

In this paper, we consider the external visibility coverage for polyhedra under the orthographic viewing model. The problem is to compute whether the whole boundary of a polyhedron is visible from a finite set of view directions, and if so, how to compute a minimal set of such view directions. A global visibility map based method is developed to calculate an optimal or near-optimal solution using object space segmentation and viewpoint space sampling. Our method subdivides the concave regions of a polyhedron into a ground set containing two types of segments: concave regions with nonempty global visibility maps and convex polygons in the concave regions whose global visibility maps are empty. The viewpoint space is sampled using a generate-as-required heuristic. The corresponding visibility matrix is computed based on global visibility calculation. Our problem is then modeled as an instance of the classical set-cover problem and solved using a minimal visible set based branch-and-bound algorithm.

A novel method for extracting the salient critical points of meshes, possibly with noise, is presented by combining mesh saliency with Morse theory. In this paper, we use the idea of mesh saliency as a measure of regional importance for meshes. The proposed method defines the salient critical points in a scalar function space using a center-surround filter operator on Gaussian-weighted average of the scalar of vertices. Compared to using a purely geometric measure of shape, such as curvature, our method yields more satisfactory results with the lower number of critical points. We demonstrate the effectiveness of this approach by comparing our results with the results of the conventional approaches in a number of examples. Furthermore, this work has a variety of potential applications. We give a direct application to the hierarchical topological representation for meshes by combining the salient critical points with the Morse-Smale complex.

This paper considers computation of the exact visibility range (or the spherical visibility map) for a closed polyhedron whose boundary is represented as a triangle mesh. For each facet on the mesh, we calculate the set of view directions from which all the points on the facet can be seen from the exterior. The projection of those visible directions onto the unit sphere forms the visibility map for the facet. We show that the exact visibility map is a spherical arrangement of closed 0-cells, 1-cells, and 2-cells embedded on the surface of the unit sphere. Based on a provable method for calculating the potential occlusion regions of a facet, a vector visibility algorithm is developed for computing the exact solution of the spherical visibility map for a facet. Examples are given to illustrate our algorithm.

In this paper, we study the problem of mesh denoising for improving the single pass surface estimation on normals and curvature tensors. We focus mainly on the engineering objects represented as dense triangle meshes. In particular, a two run non-linear diffusion algorithm based on optimal estimation theory is proposed to adaptively filter out the undesired discontinuities introduced by noise while preserving the underlying features. We show that the proposed filter can successfully improve the local surface estimates while preserving the desired features in terms of tangential and curvature discontinuities.