Munir Zaman

University Course Timetabling Using a Hybrid Harmony Search Metaheuristic Algorithm

University course timetabling problem (UCTP) is considered to be a hard combinatorial optimization problem to assign a set of events to a set of rooms and timeslots. Although several methods have been investigated, due to the nature of UCTP, memetic computing techniques have been more effective. A key feature of memetic computing is the hybridization of a population-based global search and the local improvement. Such hybridization is expected to strike a balance between exploration and exploitation of the search space. In this paper, a memetic computing technique that is designed for UCTP, called the hybrid harmony search algorithm (HHSA), is proposed. In HHSA, the harmony search algorithm (HSA), which is a metaheuristic population-based method, has been hybridized by: 1) hill climbing, to improve local exploitation; and 2) a global-best concept of particle swarm optimization to improve convergence. The results were compared against 27 other methods using the 11 datasets of Socha et al. comprising five small, five medium, and one large datasets. The proposed method achieved the optimal solution for the small dataset with comparable results for the medium datasets. Furthermore, in the most complex and large datasets, the proposed method achieved the best results.

Interval-based time synchronisation of sensor data in a mobile robot

This paper addresses the problem of time synchronisation of odometry and vision data in a mobile robot. The overall aim is to improve robot localisation by combining vision with odometry data without restricting the motion of the robot. We propose a novel method of synchronising the times of the odometry and image data using an interval paradigm, from the discrete timestamped data sets alone. The interval paradigm provides guaranteed bounds on the absolute time difference between the two data sets. A significant enhancement of the proposed method is also proposed. Simulation of the proposed method is described and the method has been successfully applied to the time synchronisation of odometry and image data in a mobile robot.

High Precision Relative Localization Using a Single Camera

In this paper a method for high precision relative localization using a single camera is proposed. The method provides pose estimates comparable in resolution to wheel odometry, but being independent of the kinematics and based on an exteroceptive sensor, is resistant to wheel slippage. The concept is based on extracting the planar transformations between frames from a sequence of ground images, which correspond to the change in robot pose. Results on a plain colored carpeted surface provide the proof of concept of the method as an alternative to wheel odometry. The contributions in this paper include a method to estimate the planar transformations between images to a high degree of precision (e.g., 0.01 degrees), and a method to calibrate the system using a ID calibration object and known motions of the robot.

A neural network-based point registration method for 3D rigid face image

Intelligent detection of human face image combined with the real-time video monitoring has been applied to improve the secure and protective possibility. The registration is an indispensible step before distinguishing the variation among the images. Neural network (NN) has a strong learning ability from a mass unstructured point cloud even containing noisy data. Neural network has been applied to register 3D reconstructed ear data and 3D surface of bunny and to achieve the better results. Motivated by this idea, NN-based registration method for 3D rigid face image is proposed. This paper presented the proof process of obtaining rotation matrix and translation vector according to the training process of neural network. Then the measure index of registration performance was provided. The elaborate experiments were conducted on the 3D USF face database (provided by the University of South Florida) to verify the effectiveness of neural network as a registration method. Next, two comparisons were performed, one group was NN-based and ICP-based registration methods and the other group was our proposed NN-based and other NN-based registration methods. The experimental results showed that neural network is a robust and potential registration method for rigid face image registration. Furthermore, our proposed NN-based registration method is extended easily to do 2D-to-3D registration and non-rigid face registration.

Localizing Pipe inspection robot using visual odometry

There is a special type of concrete pipe beneath the roads in Malaysia which is called culvert. Detecting the place of damages in these pipes is important for maintenance operations. Pipe inspection robots are one of the most reliable ways to achieve this goal. Because of the wheel slippage, low speed motion and dynamic changes in kinematic of the robot, the INS and wheel encoder methods are not accurate enough for localizing the robot inside a culvert. In this paper, we propose a solution based on monocular visual odometry. We show that although the surface of the culvert is not flat, nevertheless, by selecting an appropriate camera the optical flow of the pixels inside a small area near the center of the image is almost equal and for this reason the 3D motion of the robot can be estimated using the derivative of camera parameters. The experimental result shows this method is reliable and can be successfully used for localizing the robot inside the culverts.

Bi-Stage Large Point Set Registration Using Gaussian Mixture Models

Point set registration is to determine correspondences between two different point sets, then recover the spatial transformation between them. Many current methods, become extremely slow as the cardinality of the point set increases; making them impractical for large point sets. In this paper, we propose a bi-stage method called bi-GMM-TPS, based on Gaussian Mixture Models and Thin-Plate Splines (GMM-TPS). The first stage deals with global deformation. The two point sets are grouped into clusters independently using K-means clustering. The cluster centers of the two sets are then registered using a GMM based method. The point sets are subsequently aligned based on the transformation obtained from cluster center registration. At the second stage, the GMM based registration method is again applied, to fine tune the alignment between the two clusters to address local deformation. Experiments were conducted on eight publicly available datasets, including large point clouds. Comparative experimental results demonstrate that the proposed method, is much faster than state-of-the-art methods GMM-TPS and QPCCP (Quadratic Programming based Cluster Correspondence Projection); especially on large non-rigid point sets, such as the swiss roll, bunny and USF face datasets, and challenging datasets with topological ambiguity such as the banana dataset. Although the Coherent Point Drift (CPD) method has comparable computational speed, it is less robust than bi-GMM-TPS. Especially for large point sets, under conditions where the number of clusters is not extreme, a complexity analysis shows that bi-GMM-TPS is more efficient than GMM-TPS.

Division-based Large Point Set Registration Using Coherent Point Drift (CPD) with Automatic Parameter Tuning

Large point sets consists of unordered sets of usually 3D coordinates representing a surface (e.g., face) or a volume. With the advent of laser scanners the surface can be captured with high resolution generating a large amount of data. Processing this amount of data for point set registration efficiently, poses the type of challenges being addressed by the big data community. Coherent Point Drift (CPD) is a state-of-the-art point set registration method, that is able to handle large point cloud registration in O(n) time with the incorporation of the Fast Gauss Transform (FGT) and low-rank matrix approximation (LRA). However, its registration accuracy degrades rapidly for large point sets. To overcome this, we present a strategy that divides a large point set into several smaller overlapping subsets. These subsets are then independently registered using CPD that are then merged for final registration. To improve registration accuracy, we also propose a method to tune the width parameter of the Gaussian kernel in CPD. The proposed method has been tested on four large datasets, including the USF 3D face dataset. The results show that the proposed method is able to register large datasets with greater speed and accuracy than the state-of-the-art CPD method.

A Generic Model, and its Validation, for the Translational Systematic Errors in Synchronous Drive Robots

Abstract—Synchronous Drive Robots (SDR) are seeing increasing use as service robots in dynamic environments. Due to the changing scenery in dynamic environments, the accuracy of proprioceptive sensors such as odometry is of greater importance. This paper proposes a generic kinematic model for the translational systematic odometry error in an n-wheeled SDR (n¸3). An unexpected behaviour of SDR is the curved path when commanded to translate, which varies with wheel orientation (which changes when commanded to rotate.) This is caused by the traction force of each wheel around the centre of mass of the robot acting as a moment. There is a further odometry error due to wheel misalignment, which does not affect the path curvature, but creates a yaw. Compared to existing works, the proposed model is explicitly validated in the instance of a 3-wheeled SDR.

Prior model evaluation from Null Space Compensation perspective with application to surface reconstruction from single images

Prior model is widely applied in the area of computer vision and computer graphics. However, there is still a lack of a general theoretical scheme for evaluating the performance of the priors and a guidance for choosing suitable models. In this paper, a general scheme is proposed for linear singular problems based on the idea of Null Space Compensation. It is proved that for a linear prior model the principal directions obtained from the singular value decomposition of the model shall not be parallel to those of the system matrix determined by the problem. It is also suggested that for a nonlinear prior, higher correlation between the null space components of the estimate data based on the given prior and those of the ground truth or controlled data indicate the better suitability of the prior. The proposed evaluation scheme is demonstrated through an application to a linearized shape from shading problem, where surface shall be reconstructed from single 2D images. Both linear model and nonlinear constraints are evaluated with experiments on both synthetic images and real images. The results validate the proposed evaluation scheme and its capability for guiding in choosing a good prior model structure.