Search results
1 – 10 of 398The conventional pedestrian detection algorithms lack in scale sensitivity. The purpose of this paper is to propose a novel algorithm of self-adaptive scale pedestrian detection…
Abstract
Purpose
The conventional pedestrian detection algorithms lack in scale sensitivity. The purpose of this paper is to propose a novel algorithm of self-adaptive scale pedestrian detection, based on deep residual network (DRN), to address such lacks.
Design/methodology/approach
First, the “Edge boxes” algorithm is introduced to extract region of interests from pedestrian images. Then, the extracted bounding boxes are incorporated to different DRNs, one is a large-scale DRN and the other one is the small-scale DRN. The height of the bounding boxes is used to classify the results of pedestrians and to regress the bounding boxes to the entity of the pedestrian. At last, a weighted self-adaptive scale function, which combines the large-scale results and small-scale results, is designed for the final pedestrian detection.
Findings
To validate the effectiveness and feasibility of the proposed algorithm, some comparison experiments have been done on the common pedestrian detection data sets: Caltech, INRIA, ETH and KITTI. Experimental results show that the proposed algorithm is adapted for the various scales of the pedestrians. For the hard detected small-scale pedestrians, the proposed algorithm has improved the accuracy and robustness of detections.
Originality/value
By applying different models to deal with different scales of pedestrians, the proposed algorithm with the weighted calculation function has improved the accuracy and robustness for different scales of pedestrians.
Details
Keywords
Nima Tajallipour, Vimal Kumar and Marius Paraschivoiu
The purpose of this paper is to investigate a large‐eddy simulation, using low order numerical discretization and upwinding schemes on unstructured grids, for a turbulent free jet…
Abstract
Purpose
The purpose of this paper is to investigate a large‐eddy simulation, using low order numerical discretization and upwinding schemes on unstructured grids, for a turbulent free jet at Mach number 0.95. The accuracy and stability performance is discussed for the finite element/volume upwinding numerical code used.
Design/methodology/approach
This code is equipped with a self‐adaptive upwinding method which has been previously developed to reduce the numerical dissipation of applied low order flux calculation on unstructured elements using Roe's scheme. Herein, this method is used to numerically investigate a high Reynolds, compressible turbulent free jet and compare the results with a recently published set of experimental data. The effect of grid size is also investigated. A reasonable good agreement with the experimental measurements is obtained.
Findings
Based on the results, it is concluded that the developed self‐adaptive upwinding scheme provides a considerably better emulation of the flow regime in comparison to the full‐upwinding scheme. Different case studies have been carried out to assess the performance of self‐adaptive upwinding method and the effect of the subgrid model.
Originality/value
This paper presents an original research on self‐adaptive upwinding scheme and the effect of the subgrid model on a compressible turbulent free jet.
Details
Keywords
Binghai Zhou, Qi Yi, Xiujuan Li and Yutong Zhu
This paper aims to investigate a multi-objective electric vehicle’s (EV’s) synergetic scheduling problem in the automotive industry, where a synergetic delivery mechanism to…
Abstract
Purpose
This paper aims to investigate a multi-objective electric vehicle’s (EV’s) synergetic scheduling problem in the automotive industry, where a synergetic delivery mechanism to coordinate multiple EVs is proposed to fulfill part feeding tasks.
Design/methodology/approach
A chaotic reference-guided multi-objective evolutionary algorithm based on self-adaptive local search (CRMSL) is constructed to deal with the problem. The proposed CRMSL benefits from the combination of reference vectors guided evolutionary algorithm (RVEA) and chaotic search. A novel directional rank sorting procedure and a self-adaptive energy-efficient local search strategy are then incorporated into the framework of the CRMSL to obtain satisfactory computational performance.
Findings
The involvement of the chaotic search and self-adaptive energy-efficient local search strategy contributes to obtaining a stronger global and local search capability. The computational results demonstrate that the CRMSL achieves better performance than the other two well-known benchmark algorithms in terms of four performance metrics, which is inspiring for future researches on energy-efficient co-scheduling topics in manufacturing industries.
Originality/value
This research fully considers the cooperation and coordination of handling devices to reduce energy consumption, and an improved multi-objective evolutionary algorithm is creatively applied to solve the proposed engineering problem.
Details
Keywords
Anders Gustafsson, Delphine Caruelle and David E. Bowen
The purpose of this paper is to provide an overview of what (service) experience is and examine it using three distinct perspectives: customer experience (CX), employee experience…
Abstract
Purpose
The purpose of this paper is to provide an overview of what (service) experience is and examine it using three distinct perspectives: customer experience (CX), employee experience (EX) and human experience (HX).
Design/methodology/approach
The present conceptualization blends the marketing and organizational behavior/human resources management (OB/HRM) disciplines to clarify and reflect over the meaning of (service) experience. The marketing discipline illuminates the concept of CX, whereas the OB/HRM discipline illuminates the concept of EX. The concept of HX, which transcends CX and EX, is examined in light of its recent development in service research. For each of the three concepts, key themes are identified, and future research directions are proposed.
Findings
Because the goal that individuals seek to achieve depends on the role they are enacting, each of the three perspectives on experience (CX, EX and HX) should have a different focal point. CX requires to focus on the process of solving customer goals. EX necessitates to think in terms of organizational context and job content that support employees. Finally, the focus of HX should be on well-being via enhanced gratification, and reduced violation, of basic human needs.
Originality/value
This paper offers an interdisciplinary perspective on (service) experience and simultaneously addresses CX, EX and HX in order to reconcile the different perspectives on experience in service research.
Details
Keywords
Neetika Jain and Sangeeta Mittal
A cost-effective way to achieve fuel economy is to reinforce positive driving behaviour. Driving behaviour can be controlled if drivers can be alerted for behaviour that results…
Abstract
Purpose
A cost-effective way to achieve fuel economy is to reinforce positive driving behaviour. Driving behaviour can be controlled if drivers can be alerted for behaviour that results in poor fuel economy. Fuel consumption must be tracked and monitored instantaneously rather than tracking average fuel economy for the entire trip duration. A single-step application of machine learning (ML) is not sufficient to model prediction of instantaneous fuel consumption and detection of anomalous fuel economy. The study designs an ML pipeline to track and monitor instantaneous fuel economy and detect anomalies.
Design/methodology/approach
This research iteratively applies different variations of a two-step ML pipeline to the driving dataset for hatchback cars. The first step addresses the problem of accurate measurement and prediction of fuel economy using time series driving data, and the second step detects abnormal fuel economy in relation to contextual information. Long short-term memory autoencoder method learns and uses the most salient features of time series data to build a regression model. The contextual anomaly is detected by following two approaches, kernel quantile estimator and one-class support vector machine. The kernel quantile estimator sets dynamic threshold for detecting anomalous behaviour. Any error beyond a threshold is classified as an anomaly. The one-class support vector machine learns training error pattern and applies the model to test data for anomaly detection. The two-step ML pipeline is further modified by replacing long short term memory autoencoder with gated recurrent network autoencoder, and the performance of both models is compared. The speed recommendations and feedback are issued to the driver based on detected anomalies for controlling aggressive behaviour.
Findings
A composite long short-term memory autoencoder was compared with gated recurrent unit autoencoder. Both models achieve prediction accuracy within a range of 98%–100% for prediction as a first step. Recall and accuracy metrics for anomaly detection using kernel quantile estimator remains within 98%–100%, whereas the one-class support vector machine approach performs within the range of 99.3%–100%.
Research limitations/implications
The proposed approach does not consider socio-demographics or physiological information of drivers due to privacy concerns. However, it can be extended to correlate driver's physiological state such as fatigue, sleep and stress to correlate with driving behaviour and fuel economy. The anomaly detection approach here is limited to providing feedback to driver, it can be extended to give contextual feedback to the steering controller or throttle controller. In the future, a controller-based system can be associated with an anomaly detection approach to control the acceleration and braking action of the driver.
Practical implications
The suggested approach is helpful in monitoring and reinforcing fuel-economical driving behaviour among fleet drivers as per different environmental contexts. It can also be used as a training tool for improving driving efficiency for new drivers. It keeps drivers engaged positively by issuing a relevant warning for significant contextual anomalies and avoids issuing a warning for minor operational errors.
Originality/value
This paper contributes to the existing literature by providing an ML pipeline approach to track and monitor instantaneous fuel economy rather than relying on average fuel economy values. The approach is further extended to detect contextual driving behaviour anomalies and optimises fuel economy. The main contributions for this approach are as follows: (1) a prediction model is applied to fine-grained time series driving data to predict instantaneous fuel consumption. (2) Anomalous fuel economy is detected by comparing prediction error against a threshold and analysing error patterns based on contextual information.
Details
Keywords
Binghai Zhou, Xiujuan Li and Yuxian Zhang
This paper aims to investigate the part feeding scheduling problem with electric vehicles (EVs) for automotive assembly lines. A point-to-point part feeding model has been…
Abstract
Purpose
This paper aims to investigate the part feeding scheduling problem with electric vehicles (EVs) for automotive assembly lines. A point-to-point part feeding model has been formulated to minimize the number of EVs and the maximum handling time by specifying the EVs and sequence of all the delivery tasks.
Design/methodology/approach
First, a mathematical programming model of point-to-point part feeding scheduling problem (PTPPFSP) with EVs is presented. Because the PTPPFSP is NP-hard, an improved multi-objective cuckoo search (IMCS) algorithm is developed with novel search strategies, possessing the self-adaptive Levy flights, the Gaussian mutation and elite selection strategy to strengthen the algorithm’s optimization performance. In addition, two local search operators are designed for deep optimization. The effectiveness of the IMCS algorithm is verified by dealing with the PTPPFSP in different problem scales.
Findings
Numerical experiments are used to demonstrate how the IMCS algorithm serves as an efficient method to solve the PTPPFSP with EVs. The effectiveness and feasibility of the IMCS algorithm are validated by approximate Pareto fronts obtained from the instances of different problem scales. The computational results show that the IMCS algorithm can achieve better performance than the other high-performing algorithms in terms of solution quality, convergence and diversity.
Research limitations/implications
This study is applicable without regard to the breakdown of EVs. The current research contributes to the scheduling of in-plant logistics for automotive assembly lines, and it could be modified to cope with similar part feeding scheduling problems characterized by just-in-time (JIT) delivery.
Originality/value
Both limited electricity capacity and no earliness and tardiness constraints are considered, and the scheduling problem is solved satisfactorily and innovatively for an efficient JIT part feeding with EVs applied to in-plant logistics.
Details
Keywords
Xiaoyue Zhu, Yaoguo Dang and Song Ding
Aiming to address the forecasting dilemma of seasonal air quality, the authors design the novel self-adaptive seasonal adjustment factor to extract the seasonal fluctuation…
Abstract
Purpose
Aiming to address the forecasting dilemma of seasonal air quality, the authors design the novel self-adaptive seasonal adjustment factor to extract the seasonal fluctuation information about the air quality index. Based on the novel self-adaptive seasonal adjustment factor, the novel seasonal grey forecasting models are established to predict the air quality in China.
Design/methodology/approach
This paper constructs a novel self-adaptive seasonal adjustment factor for quantifying the seasonal difference information of air quality. The novel self-adaptive seasonal adjustment factor reflects the periodic fluctuations of air quality. Therefore, it is employed to optimize the data generation of three conventional grey models, consisting of the GM(1,1) model, the discrete grey model and the fractional-order grey model. Then three novel self-adaptive seasonal grey forecasting models, including the self-adaptive seasonal GM(1,1) model (SAGM(1,1)), the self-adaptive seasonal discrete grey model (SADGM(1,1)) and the self-adaptive seasonal fractional-order grey model (SAFGM(1,1)), are put forward for prognosticating the air quality of all provinces in China .
Findings
The experiment results confirm that the novel self-adaptive seasonal adjustment factors promote the precision of the conventional grey models remarkably. Simultaneously, compared with three non-seasonal grey forecasting models and the SARIMA model, the performance of self-adaptive seasonal grey forecasting models is outstanding, which indicates that they capture the seasonal changes of air quality more efficiently.
Research limitations/implications
Since air quality is affected by various factors, subsequent research may consider including meteorological conditions, pollutant emissions and other factors to perfect the self-adaptive seasonal grey models.
Practical implications
Given the problematic air pollution situation in China, timely and accurate air quality forecasting technology is exceptionally crucial for mitigating their adverse effects on the environment and human health. The paper proposes three self-adaptive seasonal grey forecasting models to forecast the air quality index of all provinces in China, which improves the adaptability of conventional grey models and provides more efficient prediction tools for air quality.
Originality/value
The self-adaptive seasonal adjustment factors are constructed to characterize the seasonal fluctuations of air quality index. Three novel self-adaptive seasonal grey forecasting models are established for prognosticating the air quality of all provinces in China. The robustness of the proposed grey models is reinforced by integrating the seasonal irregularity. The proposed methods acquire better forecasting precisions compared with the non-seasonal grey models and the SARIMA model.
Details
Keywords
Jie Chu, Junhong Li, Yizhe Jiang, Weicheng Song and Tiancheng Zong
The Wiener-Hammerstein nonlinear system is made up of two dynamic linear subsystems in series with a static nonlinear subsystem, and it is widely used in electrical, mechanical…
Abstract
Purpose
The Wiener-Hammerstein nonlinear system is made up of two dynamic linear subsystems in series with a static nonlinear subsystem, and it is widely used in electrical, mechanical, aerospace and other fields. This paper considers the parameter estimation of the Wiener-Hammerstein output error moving average (OEMA) system.
Design/methodology/approach
The idea of multi-population and parameter self-adaptive identification is introduced, and a multi-population self-adaptive differential evolution (MPSADE) algorithm is proposed. In order to confirm the feasibility of the above method, the differential evolution (DE), the self-adaptive differential evolution (SADE), the MPSADE and the gradient iterative (GI) algorithms are derived to identify the Wiener-Hammerstein OEMA system, respectively.
Findings
From the simulation results, the authors find that the estimation errors under the four algorithms stabilize after 120, 30, 20 and 300 iterations, respectively, and the estimation errors of the four algorithms converge to 5.0%, 3.6%, 2.7% and 7.3%, which show that all four algorithms can identify the Wiener-Hammerstein OEMA system.
Originality/value
Compared with DE, SADE and GI algorithm, the MPSADE algorithm not only has higher parameter estimation accuracy but also has a faster convergence speed. Finally, the input–output relationship of laser welding system is described and identified by the MPSADE algorithm. The simulation results show that the MPSADE algorithm can effectively identify parameters of the laser welding system.
Details
Keywords
Siyun Liu, Wenzeng Zhang and Jie Sun
Underactuated fingers are adapted to generate several grasping modes for different tasks, and coupled fingers and self-adaptive fingers are two important types of them. Aiming to…
Abstract
Purpose
Underactuated fingers are adapted to generate several grasping modes for different tasks, and coupled fingers and self-adaptive fingers are two important types of them. Aiming to expand the application and increase adaptability of robotic hand, this paper aims to propose a novel grasping model, called coupled and indirectly self-adaptive (CISA) grasping model, which is the combination of coupled finger and indirectly self-adaptive finger.
Design/methodology/approach
CISA grasping process includes two stages: first, coupled and then indirectly self-adaptive grasping; thus, it is not only integrated with the good pinching ability of coupled finger but also characterized with the high flexibility of indirectly self-adaptive finger. Furthermore, a CISA hand with linkage-slider, called CISA-LS hand, is designed based on the CISA grasping model, consisting of 1 palm, 5 CISA-LS fingers and 14 degrees of freedom.
Findings
To research the grasping behavior of CISA-LS hand, kinematic analysis, dynamic analysis and force analysis of 2-joint CISA-LS finger are performed. Results of grasping experiments for different objects demonstrate the high reliability and stability of CISA-LS hand.
Originality/value
CISA fingers integrate two grasping modes, coupled grasping and indirectly self-adaptive grasping, into one finger. And a double-linkage-slider mechanism is designed as the switch device.
Details
Keywords
Xiao‐Bing Hu, Ezequiel Di Paolo and Shu‐Fan Wu
The purpose of this paper is to present a comprehensive self‐adaptive genetic algorithm (GA) based on fuzzy mechanism, aiming to improve both the optimizing capability and the…
Abstract
Purpose
The purpose of this paper is to present a comprehensive self‐adaptive genetic algorithm (GA) based on fuzzy mechanism, aiming to improve both the optimizing capability and the convergence speed.
Design/methodology/approach
Many key factors that affect the performance of GAs are identified and analyzed, and their influences on the optimizing capability and the convergence speed are further elaborated, which prove to be very difficult to be described with explicit mathematical formulas. Therefore, a set of fuzzy rules are used to model these complicated relationships, in order to effectively guide the online self‐adaptive adjustments, such as changing the crossover and mutation probabilities, and thus to improve the optimizing capability and convergence speed.
Findings
Simulation results illustrates that, compared with a normal GA and another self‐adaptive GA based on explicit mathematical modeling of the key factors, the new GA is more advanced in terms of the optimizing capability and the convergence speed.
Originality/value
This paper develops a fuzzy‐rule‐based approach to describe the relationships between multiple GA parameters and online states, and the approach is useful in the design of a comprehensive self‐adaptive GA.
Details