Plantation monitoring and yield estimation using autonomous quadcopter for precision agriculture

Abstract

Recently, quadcopters with their advance sensors and imaging capabilities have become an imperative part of the precision agriculture. In this work, we have described a framework which performs plantation monitoring and yield estimation using the supervised learning approach, while autonomously navigating through an inter-row path of the plantation. The proposed navigation framework assists the quadcopter to follow a sequence of collision-free GPS way points and has been integrated with ROS (Robot Operating System). The trajectory planning and control module of the navigation framework employ convex programming techniques to generate minimum time trajectory between way-points and produces appropriate control inputs for the quadcopter. A new 'pomegranate dataset' comprising of plantation surveillance video and annotated frames capturing the varied stages of pomegranate growth along with the navigation framework are being delivered as a part of this work.

Fundamental Actuation Properties of Multirotors: Force–Moment Decoupling and Fail–Safe Robustness

“…3, tilting the propeller 3 of an angle α about − −−− → O B O P 3 and the propeller 6 of an angle −α about − −−− → O B O P 6 does tilt the two moments generated by the two opposite rotors in the same way and therefore keeps them collinear. The same holds for the pairs (1,4) and (2,5). As a consequence the discussion provided in Sec.…”

“…Quadrotors constitute the most common unmanned aerial vehicle (UAV) currently used in civil and industrial context. Their high versatility allows their application field to range from exploration and mapping to grasping, from monitoring and surveillance to transportiation [1]- [4] Physical interaction for quadrotors has been enabled by theoretical tools such has, e.g., physical property reshaping [5] and external wrench estimation [6]. Nevertheless, the interest of robotic communities is now moving toward modeling, design and control of more complex multi-rotor platforms, where the number of propellers is larger than four [7]- [12].…”

“…Intuitively, the intrinsic redundancy of these platforms can be exploited in order to enhance fundamental actuation properties as the possibility to independently control the position and the attitude of the vehicle and the robustness to rotor-failures which constitute key requirements for the real-world deployment. However having a redundant number of propellers is not in general enough to allow a static and safe hovering (i.e., a hovering in which both the linear and the angular velocity are zero) as it can be seen for example in [17] where experiments are shown in which the hexarotor starts to spin when control of a propeller is lost, even if still five propellers are available 1 and from other commercially available platforms 2 . A more indepth theoretical understanding of the fundamental actuation properties of multi-rotors is needed to handle those critical and extremely important situations.…”

Visual Appearance Analysis of Forest Scenes for Monocular SLAM

“…Recently, ORB-SLAM has been used on a Bebop UAV [18] for monitoring crops in plantations, combined with GPS and Inertial Measurement Unit (IMU) data for navigation.…”

An Automatic Non-Destructive Method for the Classification of the Ripeness Stage of Red Delicious Apples in Orchards Using Aerial Video

“…However, the images are usually high resolution and there is very little research in estimating the state of fruit ripening using low-resolution images, as those obtained with satellites or drones. For example, in [4], a system is presented to detect fruits in a crop of pomegranates to monitor its growth and estimate the yield, using quadcopters.…”