Improved random sampling consensus algorithm for vision navigation of intelligent harvester robot

2020 ◽  
Vol 47 (6) ◽  
pp. 881-887
Author(s):  
Bin Li ◽  
Yu Yang ◽  
Chengshuai Qin ◽  
Xiao Bai ◽  
Lihui Wang
Keyword(s):  
Real Time ◽  
Precision Agriculture ◽  
Random Sampling ◽  
Median Filter ◽  
Detection Algorithm ◽  
Consensus Algorithm ◽  
Vision Navigation ◽  
Content Type ◽  
Navigation Path ◽  
Random Sampling Consensus

Purpose Focusing on the problem that the visual detection algorithm of navigation path line in intelligent harvester robot is susceptible to interference and low accuracy, a navigation path detection algorithm based on improved random sampling consensus is proposed. Design/methodology/approach First, inverse perspective mapping was applied to the original images of rice or wheat to restore the three-dimensional spatial geometric relationship between rice or wheat rows. Second, set the target region and enhance the image to highlight the difference between harvested and unharvested rice or wheat regions. Median filter is used to remove the intercrop gap interference and improve the anti-interference ability of rice or wheat image segmentation. The third step is to apply the method of maximum variance to thresholding the rice or wheat images in the operation area. The image is further segmented with the single-point region growth, and the harvesting boundary corner is detected to improve the accuracy of the harvesting boundary recognition. Finally, fitting the harvesting boundary corner point as the navigation path line improves the real-time performance of crop image processing. Findings The experimental results demonstrate that the improved random sampling consensus with an average success rate of 94.6% has higher reliability than the least square method, probabilistic Hough and traditional random sampling consensus detection. It can extract the navigation line of the intelligent combine robot in real time at an average speed of 57.1 ms/frame. Originality/value In the precision agriculture technology, the accurate identification of the navigation path of the intelligent combine robot is the key to realize accurate positioning. In the vision navigation system of harvester, the extraction of navigation line is its core and key, which determines the speed and precision of navigation.

Innovation approach to detect the faults in multidimensional dynamic systems

Kybernetes ◽  
2010 ◽  
Vol 39 (1) ◽  
pp. 127-139 ◽  
Author(s):  
Chingiz Hajiyev ◽  
Ali Okatan
Keyword(s):  
Kalman Filter ◽  
Real Time ◽  
Dynamic Systems ◽  
Detection Algorithm ◽  
Statistical Characteristics ◽  
Sensor Faults ◽  
Content Type ◽  
Longitudinal Dynamics ◽  
Mean And Variance ◽  
The Mean

PurposeThe purpose of this paper is to design the fault detection algorithm for multidimensional dynamic systems using a new approach for checking the statistical characteristics of Kalman filter innovation sequence.Design/methodology/approachThe proposed approach is based on given statistics for the mathematical expectation of the spectral norm of the normalized innovation matrix of the Kalman filter.FindingsThe longitudinal dynamics of an aircraft as an example is considered, and detection of various sensor faults affecting the mean and variance of the innovation sequence is examined.Research limitations/implicationsA real‐time detection of sensor faults affecting the mean and variance of the innovation sequence, applied to the linearized aircraft longitudinal dynamics, is examined. The non‐linear longitudinal dynamics model of an aircraft is linearized. Faults affecting the covariances of the innovation sequence are not considered in the paper.Originality/valueThe proposed approach permits simultaneous real‐time checking of the expected value and the variance of the innovation sequence and does not need a priori information about statistical characteristics of this sequence in the failure case.

A New Method for Parameters Extraction of Extruded Surface

2011 ◽  
Vol 474-476 ◽  
pp. 834-839
Author(s):  
Luo Heng Yan ◽  
Zhong Min Huangfu
Keyword(s):  
Reverse Engineering ◽  
Random Sampling ◽  
Least Squares Method ◽  
Consensus Algorithm ◽  
Measure Data ◽  
Model Parameters ◽  
Cad Model ◽  
Parameters Extraction ◽  
Random Sampling Consensus ◽  
Crucial Parameter

The purpose of reverse engineering is to convert a large points cloud into a CAD model. Parameters extraction for extruded surface from measure data is an important problerm in reverse engineering. Extruded direction is a crucial parameter of extruded surface. In this paper, random sampling consensus algorithm combined with Least-Squares method is used to extract the extruded direction of extruded surface. Experimental results show effectiveness, robustness and accuracy of this proposed approach.

A method for real-time error detection in low-cost environmental sensors data

2019 ◽  
Vol 8 (4) ◽  
pp. 338-350
Author(s):  
Mauricio Loyola
Keyword(s):  
Real Time ◽  
Error Detection ◽  
Measurement Errors ◽  
Low Cost ◽  
Detection Algorithm ◽  
Environmental Sensors ◽  
Content Type ◽  
Building Monitoring ◽  
Learning Techniques

Purpose The purpose of this paper is to propose a simple, fast, and effective method for detecting measurement errors in data collected with low-cost environmental sensors typically used in building monitoring, evaluation, and automation applications. Design/methodology/approach The method combines two unsupervised learning techniques: a distance-based anomaly detection algorithm analyzing temporal patterns in data, and a density-based algorithm comparing data across different spatially related sensors. Findings Results of tests using 60,000 observations of temperature and humidity collected from 20 sensors during three weeks show that the method effectively identified measurement errors and was not affected by valid unusual events. Precision, recall, and accuracy were 0.999 or higher for all cases tested. Originality/value The method is simple to implement, computationally inexpensive, and fast enough to be used in real-time with modest open-source microprocessors and a wide variety of environmental sensors. It is a robust and convenient approach for overcoming the hardware constraints of low-cost sensors, allowing users to improve the quality of collected data at almost no additional cost and effort.

Real-time location estimation for indoor navigation using a visual-inertial sensor

Sensor Review ◽  
2020 ◽  
Vol 40 (4) ◽  
pp. 455-464
Author(s):  
Zhe Wang ◽  
Xisheng Li ◽  
Xiaojuan Zhang ◽  
Yanru Bai ◽  
Chengcai Zheng
Keyword(s):  
Real Time ◽  
Inertial Sensors ◽  
Inertial Sensor ◽  
Detection Algorithm ◽  
Indoor Navigation ◽  
Location Estimation ◽  
Evaluation Function ◽  
Current Frame ◽  
Content Type ◽  
Adjacent Frame

Purpose The purpose of this study is to use visual and inertial sensors to achieve real-time location. How to provide an accurate location has become a popular research topic in the field of indoor navigation. Although the complementarity of vision and inertia has been widely applied in indoor navigation, many problems remain, such as inertial sensor deviation calibration, unsynchronized visual and inertial data acquisition and large amount of stored data. Design/methodology/approach First, this study demonstrates that the vanishing point (VP) evaluation function improves the precision of extraction, and the nearest ground corner point (NGCP) of the adjacent frame is estimated by pre-integrating the inertial sensor. The Sequential Similarity Detection Algorithm (SSDA) and Random Sample Consensus (RANSAC) algorithms are adopted to accurately match the adjacent NGCP in the estimated region of interest. Second, the model of visual pose is established by using the parameters of the camera itself, VP and NGCP. The model of inertial pose is established by pre-integrating. Third, location is calculated by fusing the model of vision and inertia. Findings In this paper, a novel method is proposed to fuse visual and inertial sensor to locate indoor environment. The authors describe the building of an embedded hardware platform to the best of their knowledge and compare the result with a mature method and POSAV310. Originality/value This paper proposes a VP evaluation function that is used to extract the most advantages in the intersection of a plurality of parallel lines. To improve the extraction speed of adjacent frame, the authors first proposed fusing the NGCP of the current frame and the calibrated pre-integration to estimate the NGCP of the next frame. The visual pose model was established using extinction VP and NGCP, calibration of inertial sensor. This theory offers the linear processing equation of gyroscope and accelerometer by the model of visual and inertial pose.

scholarly journals Precise vehicle ego-localization using feature matching of pavement images

2020 ◽  
Vol 3 (2) ◽  
pp. 37-47
Author(s):  
Zijun Jiang ◽  
Zhigang Xu ◽  
Yunchao Li ◽  
Haigen Min ◽  
Jingmei Zhou
Keyword(s):  
Real Time ◽  
Intelligent Transportation System ◽  
Feature Matching ◽  
Consensus Algorithm ◽  
Feature Points ◽  
Content Type ◽  
Localization Accuracy ◽  
Vehicle Localization ◽  
Speeded Up Robust Features ◽  
Automatic Parking

Purpose Precise vehicle localization is a basic and critical technique for various intelligent transportation system (ITS) applications. It also needs to adapt to the complex road environments in real-time. The global positioning system and the strap-down inertial navigation system are two common techniques in the field of vehicle localization. However, the localization accuracy, reliability and real-time performance of these two techniques can not satisfy the requirement of some critical ITS applications such as collision avoiding, vision enhancement and automatic parking. Aiming at the problems above, this paper aims to propose a precise vehicle ego-localization method based on image matching. Design/methodology/approach This study included three steps, Step 1, extraction of feature points. After getting the image, the local features in the pavement images were extracted using an improved speeded up robust features algorithm. Step 2, eliminate mismatch points. Using a random sample consensus algorithm to eliminate mismatched points of road image and make match point pairs more robust. Step 3, matching of feature points and trajectory generation. Findings Through the matching and validation of the extracted local feature points, the relative translation and rotation offsets between two consecutive pavement images were calculated, eventually, the trajectory of the vehicle was generated. Originality/value The experimental results show that the studied algorithm has an accuracy at decimeter-level and it fully meets the demand of the lane-level positioning in some critical ITS applications.

3D fabric dynamic simulation based on Mapreduce

2015 ◽  
Vol 27 (6) ◽  
pp. 793-802 ◽  
Author(s):  
Hengliang Shi ◽  
Xiaolei Bai ◽  
Jianhui Duan
Keyword(s):  
Real Time ◽  
Collision Detection ◽  
Detection Algorithm ◽  
Calculation Model ◽  
3D Animation ◽  
Content Type ◽  
Collision Response ◽  
Simulation Based ◽  
Mass Spring Model ◽  
Mass Spring

Purpose – In cloth animation field, the collision detection of fabric under external force is very complex, and difficult to satisfy the needs of reality feeling and real time. The purpose of this paper is to improve reality feeling and real-time requirement. Design/methodology/approach – This paper puts forward a mass-spring model with building bounding-box in the center of particle, and designs the collision detection algorithm based on Mapreduce. At the same time, a method is proposed to detect collision based on geometric unit. Findings – The method can quickly detect the intersection of particle and triangle, and then deal with collision response according to the physical characteristics of fabric. Experiment shows that the algorithm improves real-time and authenticity. Research limitations/implications – Experiments show that 3D fabric simulation can be more efficiency through parallel calculation model − Mapreduce. Practical implications – This method can improve the reality feeling, and reduce calculation quantity. Social implications – This collision-detection can be used into more fields such as 3D games, aero simulation training and garments automation. Originality/value – This model and method have originality, and can be used to 3D animation, digital entertainment, and garment industry.

scholarly journals Real-time 2D–3D door detection and state classification on a low-power device

2021 ◽  
Vol 3 (5) ◽  
Author(s):  
João Gaspar Ramôa ◽  
Vasco Lopes ◽  
Luís A. Alexandre ◽  
S. Mogo
Keyword(s):  
Low Power ◽  
Real Time ◽  
Object Classification ◽  
Semantic Segmentation ◽  
Detection Algorithm ◽  
Power Device ◽  
Indoor Environments ◽  
State Classification ◽  
Segmentation Algorithms ◽  
Indoor Spaces

AbstractIn this paper, we propose three methods for door state classification with the goal to improve robot navigation in indoor spaces. These methods were also developed to be used in other areas and applications since they are not limited to door detection as other related works are. Our methods work offline, in low-powered computers as the Jetson Nano, in real-time with the ability to differentiate between open, closed and semi-open doors. We use the 3D object classification, PointNet, real-time semantic segmentation algorithms such as, FastFCN, FC-HarDNet, SegNet and BiSeNet, the object detection algorithm, DetectNet and 2D object classification networks, AlexNet and GoogleNet. We built a 3D and RGB door dataset with images from several indoor environments using a 3D Realsense camera D435. This dataset is freely available online. All methods are analysed taking into account their accuracy and the speed of the algorithm in a low powered computer. We conclude that it is possible to have a door classification algorithm running in real-time on a low-power device.

Sign in / Sign up

Export Citation Format

Share Document