Dyna-Q-based vector direction for path planning problem of autonomous mobile robots in unknown environments

Reinforcement learning (RL) is a popular method for solving the path planning problem of autonomous mobile robots in unknown environments. However, the primary difficulty faced by learning robots using the RL method is that they learn too slowly in obstacle-dense environments. To more efficiently solve the path planning problem of autonomous mobile robots in such environments, this paper presents a novel approach in which the robot’s learning process is divided into two phases. The first one is to accelerate the learning process for obtaining an optimal policy by developing the well-known Dyna-Q algorithm that trains the robot in learning actions for avoiding obstacles when following the vector direction. In this phase, the robot’s position is represented as a uniform grid. At each time step, the robot performs an action to move to one of its eight adjacent cells, so the path obtained from the optimal policy may be longer than the true shortest path. The second one is to train the robot in learning a collision-free smooth path for decreasing the number of the heading changes of the robot. The simulation results show that the proposed approach is efficient for the path planning problem of autonomous mobile robots in unknown environments with dense obstacles.     

CAR-Miner: An efficient algorithm for mining class-association rules

Building a high accuracy classifier for classification is a problem in real applications. One high accuracy classifier used for this purpose is based on association rules. In the past, some researches showed that classification based on association rules (or class-association rules – CARs) has higher accuracy than that of other rule-based methods such as ILA and C4.5. However, mining CARs consumes more time because it mines a complete rule set. Therefore, improving the execution time for mining CARs is one of the main problems with this method that needs to be solved. In this paper, we propose a new method for mining class-association rule. Firstly, we design a tree structure for the storage frequent itemsets of datasets. Some theorems for pruning nodes and computing information in the tree are developed after that, and then, based on the theorems, we propose an efficient algorithm for mining CARs. Experimental results show that our approach is more efficient than those used previously.     

A Comprehensive Approach for Estimating Hydraulic Quantities in a Multi-branched Estuarine System

Despite many previous investigations, there is no theoretically justifiable equation to determine the freshwater discharge, tidal velocity, and salinity in a complex estuary that contains several branches. In this study, the longitudinal distributions of freshwater and salinity concentration in multi-branched estuaries are investigated using a mathematical model, considering the energy balance by frictional head loss and the salt balance by diffusion and advection in a one-dimensional steady-state condition. We attempt to obtain the time-mean freshwater discharge rate quantitatively at a junction where the main flow separates into two branches. The salinity distribution along each branch is calculated by following the dispersion coefficient equation previously proposed for a single estuary. The salinity field is divided into the near-coast region and the upstream region, because in each segment the inland velocity varies according to the tidal flux. For computing the tidal velocity, we propose the Burgers equation and the Hoph-Cole transformation. The theoretical results obtained by these equations were found to be in good agreement with a set of observed data in the Red River estuary system.     

Arbitrary eigenvalue assignments for linear time-varying multivariable control systems

The problem of eigenvalue assignments for a class of linear time-varying multi-variable systems is considered. Using matrix operators and canonical transformations, it is shown that a time-varying system that is ‘lexicography-fixedly controllable’ can be made via state feedback to be equivalent to a time-invariant system whose eigenvalues are arbitrarily assignable. A simple algorithm for the design of the state feedback is provided.     

Design of reduced-order state estimators for linear time-varying multivariable systems

The design of reduced-order state estimators for linear time-varying multivariable systems is considered. Employing the concepts of matrix operators and the method of canonical transformations, this paper shows that there exists a reduced-order state estimator for linear time-varying systems that are ‘lexicography-fixedly observable’. In addition, the eigenvalues of the estimator can be arbitrarily assigned. A simple algorithm is proposed for the design of the state estimator.     

Performance evaluation of multi-hop csma/ca networks in fading environments

The performance of multi-hop CSMA/CA networks has in most cases been evaluated via simulations, or analytically using a perfect collision channel model. Using such methods, one can neither have access to the capture nature of the terminals nor evaluate the performance of the network in a fading environment. In this paper a new analytical framework that takes into account these capture and fading effects is presented. The network investigated in this paper has an infinite number of nodes which is assumed to be spatially Poisson distributed. The approach is to develop analytical expressions for the transition probabilities between states of nodes in the network by approximating these states as a Markov chain. A fast algorithm to solve these equations is also given. Numerical results show that the model agrees with simulation results for both omni-directional networks and networks employing beamforming.     

Assessment of different machine learning techniques in predicting the compressive strength of self-compacting concrete

The compressive strength of self-compacting concrete (SCC) needs to be determined during the construction design process. This paper shows that the compressive strength of SCC (CS of SCC) can be successfully predicted from mix design and curing age by a machine learning (ML) technique named the Extreme Gradient Boosting (XGB) algorithm, including non-hybrid and hybrid models. Nine ML techniques, such as Linear regression (LR), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Trees (DTR), Random Forest (RF), Gradient Boosting (GB), and Artificial Neural Network using two training algorithms LBFGS and SGD (denoted as ANN_LBFGS and ANN_SGD), are also compared with the XGB model. Moreover, the hybrid models of eight ML techniques and Particle Swarm Optimization (PSO) are constructed to highlight the reliability and accuracy of SCC compressive strength prediction by the XGB_PSO hybrid model. The highest number of SCC samples available in the literature is collected for building the ML techniques. Compared with previously published works’ performance, the proposed XGB method, both hybrid and non-hybrid models, is the most reliable and robust of the examined techniques, and is more accurate than existing ML methods (R² = 0.9644, RMSE = 4.7801, and MAE = 3.4832). Therefore, the XGB model can be used as a practical tool for engineers in predicting the CS of SCC.     

Corrosion protection of molybdate doped polypyrrole film prepared in succinic acid solution

ABSTRACT

Polypyrrole (PPy) was prepared on the mild steel substrate by electrochemical polymerisation in the solution containing pyrrole monomer and succinic acid. The mild steel surface could be passivated before and during electropolymerisation by molybdate. The morphology and structure of the PPy film were studied with SEM. The typical cauliflower structure of PPy was observed. Raman and IR spectroscopy showed that the obtained PPy was in an oxidised state. The thermal stability of PPy was investigated by the thermal gravimetric analysis, showing that PPy was stable at higher than 480°C. The electrochemical property of the PPy film was performed by open circuit potential, polarisation curves (I/E), and electrochemical impedance spectroscopy. The corrosion behaviour of mild steel (CT3) with PPy film in solution NaCl 3% was studied.

This paper is part of a supplementary issue from the 17th Asia-Pacific Corrosion Control Conference (APCCC-17).     

Efficient optimization of network protection design with <Emphasis Type="Italic">p</Emphasis>-cycles

The purpose of this paper is to consider network survivability designs that utilize the p-cycle, and to propose a novel ILP formulation for capacity design based on network fundamental cycles, as well as the available straddling links. Concepts of visible and hidden straddling links—which are essential components of the model presented herein—are also introduced. The proposed model caters for the case of joint optimization of a p-cycle network that can be solved without enumerating p-cycle candidates. In addition, the complexity of the proposed model is much less than any conventional model dealing with large size networks and suitable for the design of networks having multiple quality of protection (MQoP) service classes using mixed protection techniques.