Loop Closure Detection and SLAM in Vineyards with Deep Semantic Cues

Papadimitriou, Alexios; Kleitsiotis, Ioannis; Kostavelis, Ioannis; Mariolis, Ioannis; Giakoumis, Dimitrios; Likothanassis, Spiriden; Tzovaras, Dimitrios

doi:10.1109/icra46639.2022.9812419

Cited by 14 publications

(5 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This algorithm enhances localization accuracy in agricultural environments. The Graph-SLAM algorithm [36] improves closed-loop detection (LCD) through semantic segmentation based on grapevines. It is suitable for large-scale mapping of vineyards.…”

Section: Visual Slam In Agriculturementioning

confidence: 99%

MOLO-SLAM: A Semantic SLAM for Accurate Removal of Dynamic Objects in Agricultural Environments

Lv,

Yao,

Guo

et al. 2024

Agriculture

View full text Add to dashboard Cite

Visual simultaneous localization and mapping (VSLAM) is a foundational technology that enables robots to achieve fully autonomous locomotion, exploration, inspection, and more within complex environments. Its applicability also extends significantly to agricultural settings. While numerous impressive VSLAM systems have emerged, a majority of them rely on static world assumptions. This reliance constrains their use in real dynamic scenarios and leads to increased instability when applied to agricultural contexts. To address the problem of detecting and eliminating slow dynamic objects in outdoor forest and tea garden agricultural scenarios, this paper presents a dynamic VSLAM innovation called MOLO-SLAM (mask ORB label optimization SLAM). MOLO-SLAM merges the ORBSLAM2 framework with the Mask-RCNN instance segmentation network, utilizing masks and bounding boxes to enhance the accuracy and cleanliness of 3D point clouds. Additionally, we used the BundleFusion reconstruction algorithm for 3D mesh model reconstruction. By comparing our algorithm with various dynamic VSLAM algorithms on the TUM and KITTI datasets, the results demonstrate significant improvements, with enhancements of up to 97.72%, 98.51%, and 28.07% relative to the original ORBSLAM2 on the three datasets. This showcases the outstanding advantages of our algorithm.

show abstract

Section: Visual Slam In Agriculturementioning

confidence: 99%

MOLO-SLAM: A Semantic SLAM for Accurate Removal of Dynamic Objects in Agricultural Environments

Lv,

Yao,

Guo

et al. 2024

Agriculture

View full text Add to dashboard Cite

show abstract

“…Of course, no map of the agricultural environment is required for this approach. Furthermore, certain agricultural vehicles can perform mapping and localization using advanced techniques based on simultaneous localization and mapping (SLAM) (Zhao et al 2020;Papadimitriou et al 2022;Shu et al,2021). With the ability of map creation, path planning (Chen et al, 2020;Santos et al 2019) is also included in the agricultural vehicle, so that the vehicle has driving automation of level 3 (SAE, 2021).…”

Section: Literature Reviewmentioning

confidence: 99%

Implementation of autonomous mobile platform for agricultural task in corridor-like environments

Tobias,

Lim,

Duke

et al. 2024

Preprint

View full text Add to dashboard Cite

The role of autonomous vehicles (AVs) in assisting people is recognised and, therefore, is in constant development in numerous fields. Specifically, the ability of AV to alleviate global stressors, including an increased potential for food shortages and the decline in workers for potentially laborious work. An area where AVs developments are particularly prevalent is in agriculture. However, the few AVs being used in agriculture are often custom-built for specific purposes and require long development time as a result. This article aims to build and evaluate a versatile architecture of a mobile platform that is implemented using off-the-shelf components so that it can be transferred to any agricultural vehicle, thus reducing the development time. The research has involved investigating and incorporating various sensors, and also developing a common software module to perform the localisation, navigation and mapping particularly suited for corridor crop agricultural environment. This architecture has been integrated and implemented on a Yamaha golf cart, infusing it with purposely positioned sensors and supportive electronics to allow a Robotic Operating System (ROS) framework to gather information and control the vehicle. As the architecture is modular in nature, it can be transferred to different customised platforms. To determine the efficacy of the mobile platform, it has gone through several outdoor field trials to test the fundamentals of an agricultural AV. The evaluation demonstrates that both mapping and navigation have satisfactory results and the mobile platform remains within 5mm of the specified distance when aiming to follow the row in a vineyard. The results from these experiments demonstrate the ability of the mobile platform to successfully transform a Yamaha golf cart into an autonomous agricultural vehicle.

show abstract

“…The EU Project BACCHUS robot is a wheeled vehicle under development to harvest grapes and care for vineyards. The BACCHUS robot uses semantic segmentation of vineyard trunks for accurate robot motion estimation and consistent metric maps [7]. Our proposed navigation architecture takes the same segmentation approach, but it is used to identify discrete positions for the robot to walk to instead.…”

Section: State Of the Artmentioning

confidence: 99%

Computer-Vision Based Real Time Waypoint Generation for Autonomous Vineyard Navigation with Quadruped Robots

Milburn¹,

Gamba

Fernandes

et al. 2023

2023 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)

View full text Add to dashboard Cite

The VINUM project seeks to address the shortage of skilled labor in modern vineyards by introducing a cuttingedge mobile robotic solution. Leveraging the capabilities of the quadruped robot, HyQReal, this system, equipped with arm and vision sensors, offers autonomous navigation and winter pruning of grapevines reducing the need for human intervention. At the heart of this approach lies an architecture that empowers the robot to easily navigate vineyards, identify grapevines with unparalleled accuracy, and approach them for pruning with precision. A state machine drives the process, deftly switching between various stages to ensure seamless and efficient task completion. The system's performance was assessed through experimentation, focusing on waypoint precision and optimizing the robot's workspace for single-plant operations. Results indicate that the architecture is highly reliable, with a mean error of 21.5cm and a standard deviation of 17.6cm for HyQReal. However, improvements in grapevine detection accuracy are necessary for optimal performance. This work is based on a computer-vision-based navigation method for quadruped robots in vineyards, opening up new possibilities for selective task automation. The system's architecture works well in ideal weather conditions, generating and arriving at precise waypoints that maximize the attached robotic arm's workspace. This work is an extension of our short paper presented at the Italian Conference on Robotics and Intelligent Machines (I-RIM), 2022 [1].

show abstract

Loop Closure Detection and SLAM in Vineyards with Deep Semantic Cues

Cited by 14 publications

References 40 publications

MOLO-SLAM: A Semantic SLAM for Accurate Removal of Dynamic Objects in Agricultural Environments

MOLO-SLAM: A Semantic SLAM for Accurate Removal of Dynamic Objects in Agricultural Environments

Implementation of autonomous mobile platform for agricultural task in corridor-like environments

Computer-Vision Based Real Time Waypoint Generation for Autonomous Vineyard Navigation with Quadruped Robots

Contact Info

Product

Resources

About