scholarly journals A Low Cost Vision-Based Road-Following System for Mobile Robots

2018 ◽  
Vol 8 (9) ◽  
pp. 1635 ◽  
Author(s):  
Haojie Zhang ◽  
David Hernandez ◽  
Zhibao Su ◽  
Bo Su
Keyword(s):  
Mobile Robots ◽  
Unscented Kalman Filter ◽  
Detection System ◽  
Tracking System ◽  
Low Cost ◽  
Recognition Rate ◽  
Road Detection ◽  
The Road ◽  
Road Following ◽  
Following Error

Navigation is necessary for autonomous mobile robots that need to track the roads in outdoor environments. These functions could be achieved by fusing data from costly sensors, such as GPS/IMU, lasers and cameras. In this paper, we propose a novel method for road detection and road following without prior knowledge, which is more suitable with small single lane roads. The proposed system consists of a road detection system and road tracking system. A color-based road detector and a texture line detector are designed separately and fused to track the target in the road detection system. The top middle area of the road detection result is regarded as the road-following target and is delivered to the road tracking system for the robot. The road tracking system maps the tracking position in camera coordinates to position in world coordinates, which is used to calculate the control commands by the traditional tracking controllers. The robustness of the system is enhanced with the development of an Unscented Kalman Filter (UKF). The UKF estimates the best road borders from the measurement and presents a smooth road transition between frame to frame, especially in situations such as occlusion or discontinuous roads. The system is tested to achieve a recognition rate of about 98.7% under regular illumination conditions and with minimal road-following error within a variety of environments under various lighting conditions.

Study on the Road Rigidity Detecting System Based on LabVIEW

2012 ◽  
Vol 562-564 ◽  
pp. 1986-1989
Author(s):  
Shi Feng Yang ◽  
Chun Qia Liu ◽  
Jing Jing Xu
Keyword(s):  
Virtual Instrument ◽  
Detection System ◽  
National Standard ◽  
Road Maintenance ◽  
Single Chip ◽  
Road Detection ◽  
The Road ◽  
Development Platform ◽  
On The Road

As an important road detection system performance parameter, road rigidity was the key link and the evaluation index of the road detection system. During the construction of road infrastructure, the implementation of road worked on the hardness testing was necessary. The American National Instruments (NI)’s virtual instrument software development platform Lab VIEW was used as the system’s development platform. Through the signal collected by the pressure sensor combined with signal conditioning circuits formed by the single chip, functions of various parts were designed to analyze the relevant parameters of the road rigidity. The test data was measured and collected according to national standard methods, at the same time, virtual instrument software and related algorithms were used to analysis of the statistics data and the state of the road hardness would be detected and researched. And thus it provided an important basis for the quality of road management and road maintenance.

scholarly journals Unpaved road detection using optimized log Gabor filter banks

2021 ◽  
Author(s):  
◽  
Pooparat Plodpradista
Keyword(s):  
Gabor Filter ◽  
Detection System ◽  
Detection Accuracy ◽  
Road Detection ◽  
High Confidence ◽  
Detection Range ◽  
Unpaved Road ◽  
The Road ◽  
Look Ahead ◽  
Log Gabor

The revised unpaved road detection system (RURD) is a novel method for detecting unpaved roads in an arid environment from color imagery collected by a forward-looking camera mounted on a moving platform. The objective is to develop and validate a novel system with the ability to detect an unpaved road at a look-ahead distance up to 40 meters that does not utilize an expensive sensor, i.e., LIDAR but instead a low-cost color camera sensor. The RURD system is composed of two stages, the road region estimation (RRE) and the road model formation (RMF). The RRE stage classifies the image patches selected at 20-meter distance from the camera and labels them to either road or non-road. The classification result is used as a high confidence road area in the image, which is used in the RMF stage. The RMF stage uses log Gabor filter bank to extract road pixels that connect to the high confidence road region and generates a 3rd degree polynomial curve to represent the road model in a given image. The road model allows the system to extend the detection range from 20 meters to farther look-ahead distance. The RURD system is evaluated with two-years worth of data collection that measures both spatial and temporal precisions. The system is also benchmarked against an algorithm from Rasmussen entitled "Grouping Dominant Orientations for Ill-Structured Roads Following", which shown an average increase detection accuracy over 30 [percent].

scholarly journals Environment Monitoring for Anomaly Detection System Using Smartphones

Sensors ◽  
2019 ◽  
Vol 19 (18) ◽  
pp. 3834 ◽  
Author(s):  
Van Khang Nguyen ◽  
Éric Renault ◽  
Ruben Milocco
Keyword(s):  
Anomaly Detection ◽  
Clustering Algorithm ◽  
Detection System ◽  
Low Cost ◽  
Mean Shift ◽  
Detection Methods ◽  
The Road ◽  
Current Threshold ◽  
Grubbs Test ◽  
Outlier Tests

Currently, the popularity of smartphones with networking capabilities equipped with various sensors and the low cost of the Internet have opened up great opportunities for the use of smartphones for sensing systems. One of the most popular applications is the monitoring and the detection of anomalies in the environment. In this article, we propose to enhance classic road anomaly detection methods using the Grubbs test on a sliding window to make it adaptive to the local characteristics of the road. This allows more precision in the detection of potholes and also building algorithms that consume less resources on smartphones and adapt better to real conditions by applying statistical outlier tests on current threshold-based anomaly detection methods. We also include a clustering algorithm and a mean shift-based algorithm to aggregate reported anomalies on data to the server. Experiments and simulations allow us to confirm the effectiveness of the proposed methods.

scholarly journals Online Road Detection under a Shadowy Traffic Image Using a Learning-Based Illumination-Independent Image

Symmetry ◽  
2018 ◽  
Vol 10 (12) ◽  
pp. 707 ◽  
Author(s):  
Yongchao Song ◽  
Yongfeng Ju ◽  
Kai Du ◽  
Weiyu Liu ◽  
Jiacheng Song
Keyword(s):  
Feature Fusion ◽  
Detection System ◽  
Geometric Mean ◽  
Detection Algorithm ◽  
Detection Methods ◽  
Support Vector ◽  
Svm Classifier ◽  
Light Illumination ◽  
Road Detection ◽  
The Road

Shadows and normal light illumination and road and non-road areas are two pairs of contradictory symmetrical individuals. To achieve accurate road detection, it is necessary to remove interference caused by uneven illumination, such as shadows. This paper proposes a road detection algorithm based on a learning and illumination-independent image to solve the following problems: First, most road detection methods are sensitive to variation of illumination. Second, with traditional road detection methods based on illumination invariability, it is difficult to determine the calibration angle of the camera axis, and the sampling of road samples can be distorted. The proposed method contains three stages: The establishment of a classifier, the online capturing of an illumination-independent image, and the road detection. During the establishment of a classifier, a support vector machine (SVM) classifier for the road block is generated through training with the multi-feature fusion method. During the online capturing of an illumination-independent image, the road interest region is obtained by using a cascaded Hough transform parameterized by a parallel coordinate system. Five road blocks are obtained through the SVM classifier, and the RGB (Red, Green, Blue) space of the combined road blocks is converted to a geometric mean log chromatic space. Next, the camera axis calibration angle for each frame is determined according to the Shannon entropy so that the illumination-independent image of the respective frame is obtained. During the road detection, road sample points are extracted with the random sampling method. A confidence interval classifier of the road is established, which could separate a road from its background. This paper is based on public datasets and video sequences, which records roads of Chinese cities, suburbs, and schools in different traffic scenes. The author compares the method proposed in this paper with other sound video-based road detection methods and the results show that the method proposed in this paper can achieve a desired detection result with high quality and robustness. Meanwhile, the whole detection system can meet the real-time processing requirement.

Path planning for mobile robots navigation with obstacle avoidance based on octrees

2020 ◽  
Vol 25 (4) ◽  
pp. 25-30
Author(s):  
Rud V.V. ◽  
◽  
Panaseiko H.N. ◽  
Keyword(s):  
Mobile Robot ◽  
Mobile Robots ◽  
Real Time ◽  
Detection System ◽  
Test Results ◽  
Depth Sensing ◽  
The Road ◽  
Existing Problems ◽  
Laser Scanners ◽  
Motor Actions

The article considers the problem of navigating mobile robots and finding the best way to the goal in real-time in a space surrounded by unknown objects. The motor actions of the robot must be defined and adapted to changes in the environment. When using only laser scanners on mobile work, objects above or below the lasers' level will remain obstacles to the robot. Current algorithms and principles of navigation are considered. Extended the existing real-time interference detection system using lasers by adding a camera that calculates the length of objects. The new system has been successfully implemented and tested in a mobile robot, ensuring the passage of the road providing collision-free paths. The obtained simulation results are presented in the article. The existing problems of navigation of mobile robots, which are moving in the particular area from their position to the specified destination on the map, were investigated. The current problem is the inability to spot objects that are not on the same level as the mobile robot's lasers. Moreover, the task is complicated when you need to recognize such objects while the robot is moving in real time. The current algorithms and principles of navigation given by previous research and publications are analyzed. As a result of the work, the existing system of recognition and avoidance of obstacles was expanded. Prior to that, the system used only odometry and information obtained from laser scanners, without obtaining data from other sources of environmental information. The idea of development was to use a camera, which was already part of the components of the researched mobile robot. It has become possible to generate a pointcloud relative to the environment, using a depth sensing camera to calculate the distance to objects. Because the density of the received data in the form of a pointcloud is too high for further processing, a downsample VoxelGrid filter was used, which reduces the density of the point cloud. VoxelGrid belongs to the PCL library. Another problem was the removal of information about unnecessary objects in the camera's field of view. These include the floor, ceiling, parts of the robot (such as a manipulator). The PassThrough filter from the PCL library was used to solve this problem. The next step is to process the filtered data using OctoMap. As a result, an octree is generated. A top-down projection is created from the octree generated in the previous step. The resulting projection must be processed and converted into polygonal obstacles. Only then they will be marked by teb_local_planner as obstacles. The developed system was successfully implemented and tested both in the Gazebo simulation and in the researche mobile robot. The path with obstacles will be completed without collisions. The paper presents the obtained test results.

scholarly journals Performance Analysis of Low-Cost Tracking System for Mobile Robots

Machines ◽  
2020 ◽  
Vol 8 (2) ◽  
pp. 29
Author(s):  
Andrea Botta ◽  
Giuseppe Quaglia
Keyword(s):  
Performance Analysis ◽  
Mobile Robots ◽  
Moving Objects ◽  
Tracking System ◽  
Low Cost ◽  
Simple Extension ◽  
Preliminary Study ◽  
Indoor Tracking ◽  
User Friendly ◽  
Straightforward Approach

This paper proposes a reliable and straightforward approach to mobile robots (or moving objects in general) indoor tracking, in order to perform a preliminary study on their dynamics. The main features of this approach are its minimal and low-cost setup and a user-friendly interpretation of the data generated by the ArUco library. By using a commonly available camera, such as a smartphone one or a webcam, and at least one marker for each object that has to be tracked, it is possible to estimate the pose of these markers, with respect to a reference conveniently placed in the environment, in order to produce results that are easily interpretable by a user. This paper presents a simple extension to the ArUco library to generate such user-friendly data, and it provides a performance analysis of this application with static and moving objects, using a smartphone camera to highlight the most notable feature of this solution, but also its limitations.

Driver Gaze Tracking and Eyes Off the Road Detection System

2015 ◽  
Vol 16 (4) ◽  
pp. 2014-2027 ◽  
Author(s):  
Francisco Vicente ◽  
Zehua Huang ◽  
Xuehan Xiong ◽  
Fernando De la Torre ◽  
Wende Zhang ◽  
...  
Keyword(s):  
Detection System ◽  
Road Detection ◽  
Gaze Tracking ◽  
The Road

scholarly journals Evaluating the Safety Risk of Rural Roadsides Using a Bayesian Network Method

2019 ◽  
Vol 16 (7) ◽  
pp. 1166 ◽  
Author(s):  
Tianpei Tang ◽  
Senlai Zhu ◽  
Yuntao Guo ◽  
Xizhao Zhou ◽  
Yang Cao
Keyword(s):  
Bayesian Network ◽  
Low Cost ◽  
Road Networks ◽  
Access Point ◽  
Safety Risk ◽  
Crash Data ◽  
Risk Levels ◽  
The Road ◽  
Run Off ◽  
Point Density

Evaluating the safety risk of rural roadsides is critical for achieving reasonable allocation of a limited budget and avoiding excessive installation of safety facilities. To assess the safety risk of rural roadsides when the crash data are unavailable or missing, this study proposed a Bayesian Network (BN) method that uses the experts’ judgments on the conditional probability of different safety risk factors to evaluate the safety risk of rural roadsides. Eight factors were considered, including seven factors identified in the literature and a new factor named access point density. To validate the effectiveness of the proposed method, a case study was conducted using 19.42 km long road networks in the rural area of Nantong, China. By comparing the results of the proposed method and run-off-road (ROR) crash data from 2015–2016 in the study area, the road segments with higher safety risk levels identified by the proposed method were found to be statistically significantly correlated with higher crash severity based on the crash data. In addition, by comparing the respective results evaluated by eight factors and seven factors (a new factor removed), we also found that access point density significantly contributed to the safety risk of rural roadsides. These results show that the proposed method can be considered as a low-cost solution to evaluating the safety risk of rural roadsides with relatively high accuracy, especially for areas with large rural road networks and incomplete ROR crash data due to budget limitation, human errors, negligence, or inconsistent crash recordings.

Sign in / Sign up

Export Citation Format

Share Document