Dayou Li

Effects of polarization on four-beam laser interference lithography

This paper demonstrates that polarization plays an important role in the formation of interference patterns, pattern contrasts, and periods in four-beam interference lithography. Three different polarization modes are presented to study the effects of polarization on four-beam laser interference based on theoretical analysis, simulations, and experiments. A four-beam laser interference system was set up to modify the silicon surface. It was found that the secondary periodicity or modulation was the result of the misaligned or unequal incident angles only in the case of the TE-TE-TM-TM mode. The resulting patterns have shown a good correspondence with the theoretical analysis and simulations.

Towards robust personal assistant robots: Experience gained in the SRS project

SRS is a European research project for building robust personal assistant robots using ROS (Robotic Operating System) and Care-O-bot (COB) 3 as the initial demonstration platform. In this paper, experience gained while building the SRS system is presented. A main contribution of the paper is the SRS autonomous control framework. The framework is divided into two parts. First, it has an automatic task planner, which initialises actions on the symbolic level. The planner produces proactive robotic behaviours based on updated semantic knowledge. Second, it has an action executive for coordination actions at the level of sensing and actuation. The executive produces reactive behaviours in well-defined domains. The two parts are integrated by fuzzy logic based symbolic grounding. As a whole, they represent the framework for autonomous control. Based on the framework, several new components and user interfaces are integrated on top of COBs existing capabilities to enable robust fetch and carry in unstructured environments. The implementation strategy and results are discussed at the end of the paper.

Superhydrophobic dual micro- and nanostructures fabricated by direct laser interference lithography

Abstract. A method for the fabrication of highly ordered superhydrophobic dual micro- and nanostructures on silicon by direct laser interference lithography (LIL) is presented. The method offers its innovation that the superhydrophobic dual micro- and nanostructures can be fabricated directly by controlling the process of four-beam laser interference and the use of hydrofluoric acid (HF) to wipe off the silica generated during the process. Different laser fluences, exposure durations, and cleanout times have been investigated to obtain the optimum value of the contact angle (CA). The superhydrophobic surface with the CA of 153.2 deg was achieved after exposure of 60 s and immersion in HF with a concentration of 5% for 3 min. Compared with other approaches, it is a facile and efficient method with its significant feature for the macroscale fabrication of highly ordered superhydrophobic dual micro- and nanostructures on silicon.

Both antireflection and superhydrophobicity structures achieved by direct laser interference nanomanufacturing

Inspired by nature, a number of techniques have been developed to fabricate the bionic structures of lotus leaves and moth eyes in order to realize the extraordinary functions of self-cleaning and antireflection. Compared with the existing technologies, we present a straightforward method to fabricate well-defined micro and nano artificial bio-structures in this work. The proposed method of direct laser interference nanomanufacturing (DLIN) takes a significant advantage of high efficiency as only a single technological procedure is needed without pretreatment, mask, and pattern transfer processes. Meanwhile, the corresponding structures show both antireflection and superhydrophobicity properties simultaneously. The developed four-beam nanosecond laser interference system configuring the TE-TE-TE-TE and TE-TE-TE-TM polarization modes was set up to generate periodic micro cone and hole structures with a huge number of nano features on the surface. The theoretical and experimental results have shown that the periodic microcone structure exhibits excellent properties with both a high contact angle (CA = 156.3°) and low omnidirectional reflectance (5.9–15.4%). Thus, DLIN is a novel and promising method suitable for mass production of self-cleaning and antireflection surface structures.

Genetic algorithm based solution to dead-end problems in robot navigation

In robot navigation, mobile robots can suffer from dead-end problems, that is, they can be stuck in areas which are surrounded by obstacles. Attempts have been reported to avoid a robot entering into such a dead-end area. However, in some applications, for example, rescue work, the dead-end areas must be explored. Therefore, it is vital for the robot to come out from the dead-end areas after exploration. This paper presents an approach which enables a robot to come out from dead-end areas. There are two main parts: a dead-end detection mechanism and a genetic algorithm (GA) based online training mechanism. When the robot realises that it is stuck in a dead-end area, it will operate the online training to produce a new best chromosome that will enable the robot to escape from the area.

Fuzzy control of a three-tank system

Abstract Fuzzy inverse reasoning performed using fuzzy relational equations can be employed to deduce control actions appropriate for a desired process output. The three-tank plant to be controlled has two inputs and hence the fuzzy relational equation describing the dynamics of the plant is a two-decision variable equation. Algorithms are only available for solving equations that have a single decision variable. A method is proposed in this paper to decompose the two-decision-variable fuzzy relational equation into single-decision-variable equations so that existing algorithms can be applied to produce control actions for the plant.

Towards automated task planning for service robots using semantic knowledge representation

Automated task planning for service robots faces great challenges in handling dynamic domestic environments. Classical methods in the Artificial Intelligence (AI) area mostly focus on relatively structured environments with fewer uncertainties. This work proposes a method to combine semantic knowledge representation with classical approaches in AI to build a flexible framework that can assist service robots in task planning at the high symbolic level. A semantic knowledge ontology is constructed for representing two main types of information: environmental description and robot primitive actions. Environmental knowledge is used to handle spatial uncertainties of particular objects. Primitive actions, which the robot can execute, are constructed based on a STRIPS-style structure, allowing a feasible solution (an action sequence) for a particular task to be created. With the Care-O-Bot (CoB) robot as the platform, we explain this work with a simple, but still challenging, scenario named “get a milk box”. A recursive back-trace search algorithm is introduced for task planning, where three main components are involved, namely primitive actions, world states, and mental actions. The feasibility of the work is demonstrated with the CoB in a simulated environment.

K-Order Surrounding Roadmaps Path Planner for Robot Path Planning

Probabilistic roadmaps are commonly used in robot path planning. Most sampling-based path planners often produce poor-quality roadmaps as they focus on improving the speed of constructing roadmaps without paying much attention to the quality. Poor-quality roadmaps can cause problems such as poor-quality paths, time-consuming path searching and failures in the searching. This paper presents a K-order surrounding roadmap (KSR) path planner which constructs a roadmap in an incremental manner. The planner creates a tree while answering a query, selects the part of the tree according to quality measures and adds the part to an existing roadmap which is obtained in the same way when answering the previous queries. The KSR path planner is able to construct high-quality roadmaps in terms of good coverage, high connectivity, provision of alternative paths and small size. Comparison between the KSR path planner and Reconfigurable Random Forest (RRF), an existing incremental path planner, as well as traditional probabilistic roadmap (PRM) path planner shows that the roadmaps constructed using the KSR path planner have higher quality that those that are built by the other planners.

Integration of Symbolic Task Planning into Operations within an Unstructured Environment

To ensure a robot capable of robust task execution in unstructured environments, task planners need to have a high-level understanding of the nature of the world, reasoning for deliberate actions, and reacting to environment changes. Proposed is a practical task planning approach that seamlessly integrating deeper domain knowledge, real time perception and symbolic planning for robot operation. A higher degree of autonomy under unstructured environment will be endowed to the robot with the proposed approach.

Identifying Mubasher software products through sentiment analysis of Arabic tweets

Social media has recently become a rich resource in mining user sentiments. In this paper, Twitter has been chosen as a platform for opinion mining in trading strategy with Mubasher products, which is a leading stock analysis software provider in the Gulf region. This experiment proposes a model for sentiment analysis of Saudi Arabic (standard and Arabian Gulf dialect) tweets to extract feedback from Mubasher products. A hybrid of natural language processing and machine learning approaches on building models are used to classify tweets according to their sentiment polarity into one of the classes positive, negative and neutral. Firstly, documents Pre-processing are explored on the dataset. Secondly, Naive Bayes and Support Vector Machines (SVMs) are applied with different feature selection schemes like TF-IDF (Term Frequency-Inverse Document Frequency) and BTO (Binary-Term Occurrence). Thirdly, the proposed model for sentiment analysis is expanded to obtain the results for N-Grams term of tokens. Finally, human has labelled the data and this may involve some mistakes in the labelling process. At this moment, neutral class with generalisation of our classification will take results to different classification accuracy.