The Auxiliary System of Cleaning Vehicle Based on Road Recognition Technology

Feng, Jiaming; Zhao, Feng’an; Ye, Meng; Sun, Wenchao

doi:10.4271/2021-01-0245

Cited by 4 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, such a method based on the response recognition of tires is affected by many external uncertainties due to the complexity of the generation mechanism of the tire noise, and sometimes it is insurmountable to accurately identify the adhesion coefficients. Recently, some identification methods on the basis of visual information have been proposed in the current study [ 25 , 26 , 27 , 28 ]. For example, given the nondeterminacy of kinematic models and deep-learning models, an image-based fusion estimation method by virtue of the virtual sensing theory was put forward to exactly realize the identification of the road surface condition in reference [ 25 ].…”

Section: Introductionmentioning

confidence: 99%

Adhesion Coefficient Identification of Wheeled Mobile Robot under Unstructured Pavement

Zhang,

Song,

et al. 2024

Sensors

View full text Add to dashboard Cite

Because of its uneven and large slope, unstructured pavement presents a great challenge to obtaining the adhesion coefficient of pavement. An estimation method of the peak adhesion coefficient of unstructured pavement on the basis of the extended Kalman filter is proposed in this paper. The identification accuracy of road adhesion coefficients under unstructured pavement is improved by introducing the equivalent suspension model to optimize the calculation of vertical wheel load and modifying vehicle acceleration combined with vehicle posture data. Finally, the multi-condition simulation experiments with Carsim are conducted, the estimation accuracy of the adhesion coefficient is at least improved by 3.6%, and then the precision and effectiveness of the designed algorithm in the article are verified.

show abstract

Section: Introductionmentioning

confidence: 99%

Adhesion Coefficient Identification of Wheeled Mobile Robot under Unstructured Pavement

Zhang,

Song,

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…CNN, including the convolution layer and pool layer, is an efficient recognition method developed in recent years. In the 1960s, Hubel and Wiesel found that the unique network structure of neurons in the cat skin layer for local sensitivity and direction selection can effectively reduce the complexity of the feedback neural network [1]. Then CNN with convolution calculation and a deep structure was proposed.…”

Section: Introductionmentioning

confidence: 99%

An Application of Machine Learning Algorithms on the Finger Image Prediction

Liu

2022

SHS Web Conf.

View full text Add to dashboard Cite

In recent years, using machine learning (ML) algorithm to analyze a picture, obtain its features, and finally identify what the picture is about is getting more and more important. This is because with the popularity of the electronic equipment and high-performance computing equipment, people began to pursue a more convenient and automatic life. The science of the image recognition frees people’s hands to a certain extent through the training of algorithms, thus making people’s life more convenient. This paper presents a comparison of two ML algorithms: Multi-layer Perceptron (MLP), and Convolutional Neural Network (CNN) with three different optimization methods on the data-set by measuring their test accuracy and their running time. The said data-set consists of a training-set of 1080 pictures (64 by 64 pixels) of signs representing numbers from 0 to 5 (180 pictures per number) and a test set of 120 pictures (64 by 64 pixels) of signs representing numbers from 0 to 5 (20 pictures per number). For the implementation of the ML algorithms, the data-set was partitioned in the following fashion: 90% for training phase, and 10% for testing phase. The hyper-parameters used for all the classifiers were manually assigned. Results show that most of the presented ML algorithms performed not bad with a test accuracy over 80%, and the CNN algorithm performed best among all the implemented algorithms with a test accuracy about 91.04%.

show abstract