An entry-exit path planner for an autonomous tractor in a paddy field

Jeon, Chan-Woo; Kim, Hak-Jin; Yun, Chang-Ho; Gang, Minseok; Han, Xiongzhe

doi:10.1016/j.compag.2021.106548

Cited by 17 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To resolve this issue, the creation of effective migration plans based on current supply-demand information and machinery operation statistics is necessary. For example, some works [4,5] have calculated the operational costs of different in-field and out-of-field activities, so that effective migration plans can be developed to optimize cross-regional farm machinery services.…”

Section: Introductionmentioning

confidence: 99%

“…Moreover, in the entire operation classification process, field-road operation classification that automatically distinguishes in-field and out-of-field activities serves as the first step. Many studies [4,5] carry out the field-road operation classification relying on an existing field boundary database; however, in China, there is no available field boundary database, making it necessary to develop a field-road operation classification method that can function without the input of field boundaries for assisting in the calculation of agricultural machines' operation statistics.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Field–Road Operation Classification of Agricultural Machine GNSS Trajectories Using Spatio-Temporal Neural Network

Chen

Zhou³

et al. 2023

Agronomy

View full text Add to dashboard Cite

The classification that distinguishes whether machines are driving on roads or working in fields based on their global navigation satellite system (GNSS) trajectories is essential for effective management of cross-regional agricultural machinery services in China. In this paper, a novel field–road classification method utilizing multiple deep neural networks (MultiDNN) is proposed to enhance the accuracy of field and road point classification. The MultiDNN model incorporates a bi-directional long short-term memory network (BiLSTM), a topology adaptive graph convolution network (TAG), and a self-attention network (ATT) to effectively extract spatio-temporal features for field–road classification. The BiLSTM is used to capture temporal relationships along the time axis of a trajectory, providing global contextual information for each point. Then, the TAG network is used to obtain the spatio-temporal relationships between adjacent points in a trajectory, offering local contextual information for each point. Finally, the ATT network assigns varying weights to features to emphasize important characteristics. The performance of the MultiDNN model was evaluated using a wheat harvesting trajectory dataset, and the results showed that it achieved a high degree of accuracy, up to 89.75%, outperforming the best baseline method (GCN) by 2.79%.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Field–Road Operation Classification of Agricultural Machine GNSS Trajectories Using Spatio-Temporal Neural Network

Chen

Zhou³

et al. 2023

Agronomy

View full text Add to dashboard Cite

show abstract

“…In recent years, machine vision and image processing technology has developed rapidly, and it has begun to be applied in various aspects such as field agricultural machinery navigation, variable spraying and weed identification (He et al, 2021;Yun et al, 2021;Jeon et al, 2021;Li et al, 2021). However, the application of this technology in the early wheat field navigation is less and not suitable, mainly because the wheat field environment is complex and diverse: (1) variations in lighting due to time differences and weather changes; (2) the shape of wheat seedlings in different growth periods was different; (3) the wheat field was affected by weeds, leaves and other debris; (4) shadows caused by agricultural machinery, power poles, and trees; (5) in the process of agricultural machinery operation, there is a certain yaw angle in the camera during the image acquisition process due to uneven ground or mechanical vibration. Therefore, how to provide a visual navigation line with strong robustness and high accuracy in a complex environment has become the hot sport of current research.…”

Section: Introductionmentioning

confidence: 99%

Experimental Study on Navigation for Wheat Seedling Root Cutting Based on Deep Learning

LIN,

XU,

2023

INMATEH

View full text Add to dashboard Cite

In response to the automatic extraction of navigation lines for wheat root cutting, this paper conducted field experiments and analyses on the navigation line extraction algorithm, based on the improved YOLOv5 algorithm. Firstly, based on the characteristics of wheat seedling rows during the wheat rejuvenation period, the YOLOv5 algorithm was improved by using rotation detection box labels, and navigation lines were extracted by fitting the detection boxes using clustering methods. Then, an experimental system was established to conduct field experiments on the algorithm: (1) Tests were conducted at three speeds of 0.5 m/s, 1.0 m/s and 1.5 m/s respectively, and the position error of the root cutter was measured and analyzed, indicating that the actual navigation path position error increased with the speed. The best navigation performance was observed at 1 m/s, with an average positional error of 18.56 mm, meeting the requirements for wheat root cutting. (2) Robustness analysis of the algorithm was conducted using data collected from 2019 to 2022. Comparative tests were conducted from four aspects: different years, different time periods, different environments, and different yaw angles. The results showed that the algorithm proposed in this paper has stronger robustness and higher accuracy.

show abstract

“…Han et al did not automate the connection path between subfields. Jeon et al [14] introduced the concept of an entry-exit path planning technique. Their method focuses on generating a continuous coverage path, the path from the entrance to the starting point, and the path from the end to the exit location of a field.…”

Section: Introductionmentioning

confidence: 99%

“…The As explained in Section 2.3.6, each subfield has 4 possible starting points on the outermost headland path, each with their own pre-determined end points along the border of the subfield. Shen et al [14] explored something like this, except they assumed that each field only had a single entrance/exit, and their connecting paths were defined as the straight-line distance between the subfield's entrance/exit. Our study aims to expand on this concept by finding the best subfield traversal order and corresponding subfield starting/ending points.…”

mentioning

confidence: 99%

Optimal Path Generation with Obstacle Avoidance and Subfield Connection for an Autonomous Tractor

et al. 2022

View full text Add to dashboard Cite

As autonomous tractors become more common crop harvesting applications, the need to optimize the global servicing path becomes crucial for maximizing efficiency and crop yield. In recent years, several methods of path generation have been researched, but very few have studied their applications on complex field shapes. In this study, a method of creating the optimal servicing path for simple and complex field shapes is proposed. The proposed algorithm creates subfields for a target land, optimizes the track direction for several subfields individually, merges subfields that result in overall increased efficiency, and finds the minimum non-operating paths to travel from subfield to subfield while selecting the respective optimal subfield starting locations. Additionally, it is required that this process must be done within 3 seconds to meet performance requirements. Results from 3 separate field shapes show that the field traversal efficiency can range from 68.0% to 94.4%, and the coverage ratio can range from 98.8% to 99.9% for several different conditions. In comparison with previous studies using the same field shape, the proposed methods demonstrate an increase of 5.5% in field traversal efficiency.

show abstract

An entry-exit path planner for an autonomous tractor in a paddy field

Cited by 17 publications

References 33 publications

Field–Road Operation Classification of Agricultural Machine GNSS Trajectories Using Spatio-Temporal Neural Network

Field–Road Operation Classification of Agricultural Machine GNSS Trajectories Using Spatio-Temporal Neural Network

Experimental Study on Navigation for Wheat Seedling Root Cutting Based on Deep Learning

Optimal Path Generation with Obstacle Avoidance and Subfield Connection for an Autonomous Tractor

Contact Info

Product

Resources

About