The world’s population is growing at an alarming rate. It is estimated that by 2050, the world and its food production will have to support more than 9 billion people! In order to make such a future sustainable, the current global food production must be doubled. However, when we quickly swallow all arable land, how should agriculture cope with this huge change?

From the group A robotics and IoT laboratory, intelligent precision agriculture and forestry in INESC TEC-technology and science use to create a ROS stack solution Husky UGV precision field crop agriculture to combat this problem. The Institute of Systems and Computer Engineering, Technology and Science is the research and development institution of the University of Porto in Portugal, it is the interface between academia and industry, and serves public administration. The team’s agricultural robotics laboratory is dedicated to developing solutions based on robotics, automation and the Internet of Things to improve the intelligent precision of agriculture and forestry in three main environments ("right time, right quantity, right location") Level, profitability and automation level:

1. Permanent crops (such as steep slope vineyards, olive trees, fruit trees)

2. Forest biomass harvest

3. Conservation cultivation (conventional and urban)

Agricultural automation

Although the agricultural industry has worked hard to include robotic automation in many of its processes, certain environments and technologies (such as planting vines and olive trees on mountainous or steep slopes) still pose major challenges to mechanization due to difficult terrain conditions. Extreme fragmentation, harsh weather conditions, a large amount of biodiversity and so on. However, in order to meet this growing demand, farmers are looking for cost-effective, safe and autonomous precision spraying/mowing robots that can work in steep vineyards to reduce costs, increase efficiency and reduce agricultural impact on the environment. Destruction.

Husky monitors the vineyard

Therefore, the AgRobIT project based on Husky UGV proposes a simplified architecture based on a ROS stack solution, which consists of three main ROS software packages: localization and mapping, path planning and control, and task manager. In the past four years, AgRobIT has been able to take advantage of the main development achievements of INESCTEC, including VineSLAM (a cost-effective, modular and reliable localization system that can use/not use GNSS and consider semantic information), and AgRobPP (used to perform Path planning and control system) to understand soil compaction and the center of gravity of the robot). However, the codes of VineSLAM and AgRobPP are closed source codes, and the team hopes to create supporting documents to make them open to the general community.

Therefore, AgRobIT proposed an open source ROS stack with advanced agricultural path planning and localized ROS software package, which considers the limitations of steep slope viticulture, and can interoperate with the general standards of agricultural decision support systems. The team believes that the AgRobIT stack will become a key supporting technology for the development of cost-effective lawn mower robots.

Click to watch the video: AgRobIT在树林中的行动

In order to provide their ROS software stack, the team includes a 3D LIDAR for positioning in agricultural terrain through laser ranging in a highly unstructured environment; a stereo camera for using natural features (such as vines) Trunk) for positioning; IMU (redundant sensor); and GNSS receiver (redundant sensor for positioning). To add manipulation (harvesting and trimming) functions, the team integrated ROBOTIS' Dynamixel Pro robot arm (Manipulator-H) with a USB-485 adapter connected to a separate processing unit. The system has its own battery and communicates with the rest of the system via Ethernet. At the ROS level, they developed a driver node to subscribe to joint position targets and publish each joint state. Most importantly, they also added their own joint controller and path planner ROS node, which has been benchmarked against other MoveIt solutions. Using its own software and onboard hardware, the platform was tested for vineyard monitoring, spraying, harvesting, pruning and pruning tasks.

Husky UGV solves the grape problem

The Husky UGV platform provides vital help for the robotic laboratory to test its concepts in various agricultural scenarios to fine-tune its own code. It allows them to focus more time and resources on creating a ROS stack instead of having to conceive a robotic platform from scratch. Due to its wide variety of software supporting ROS and its success in the field of outdoor robots, the team chose the Clearpath solution. In the end, the team chose Husky UGV because of its robustness, simplicity and modularity, and most importantly its ROS compatibility. As Filipe Neves dos Santos, Senior Researcher of Agricultural and Forestry Robots and Internet of Things, said: "We chose Husky UGV because it is compatible with ROS and is modular.

The AgRobIT team conducts on-site testing

But of course the work is not over yet. The AgRobIT team plans to launch a full-featured stack for use with ROS 2.0 to enable autonomous navigation and reliable/reliable support of GNSS in agricultural applications such as vineyards and olive ditch. In the next few years, they will fine-tune the research through further tests and continue to publish research results to help the agricultural industry adapt to technological advancements through key industry partnerships.

The Intelligent Precision Agriculture and Forestry Robotics and IoT Laboratory, founded and managed by Filipe Santos, consists of a team of 9 researchers: Hebér Sobreira, Luss Santos, Sandro Magalhães, André Aguiar, Daniel Silva, Pedro Moura, André Bianchi, Jorge Pinto and Luis · Castro.

  To learn more about husky, please click on the link.

http://www.jingtianrobots.com/index.php?id=9