VINBOT - Autonomous Mobile Robot
VINBOT is an all-terrain autonomous mobile robot with a set of sensors capable of Capturing and analysing vineyard images and 3D data by means of cloud computing applications, to determine the yield of vineyards and to share information with the winegrowers.
VINBOT responds to a need to boost the quality of European wines by implementing precision viticulture (PV) to estimate the yield (amount of fruit per square metre of vine area: kg/m2).
Wineries and wine growers will be able to make accurate yield predictions to organise the production and marketing their wines, coordinating the mixing of grapes of homogeneous quality to efficiently market a range of wines by quality and price.
VINBOT is an all-terrain autonomous mobile robot equipped with a set of sensors capable of capturing and analysing vineyard images and 3D data by means of cloud computing applications. This Precision Viticulture tool will allow winegrowers to accurately assess the yield and leave behind current estimates based on visual inspection of small samples which lead to error and low quality wines.
VINBOT will estimate the amount of leaves, grapes and other phyto-data in the vine throughout the entire vineyard via computer vision and other sensors and will generate online yield and vigour maps to help winegrowers optimise management strategies.
Winegrowers will be able to blend grapes with the same ripeness state and thus producing high quality wines. Furthermore, they will be able to segment their production in different qualities and quantities to maximise its benefits.
The VINBOT consist of:
- A robotic platform: durable, mobile, with open-source software
- Colour cameras to take high-precision images of the vine
- 3D range finders to navigate the field and to obtain the shape of the canopies
- A Normalised Difference Vegetation Index sensor to compute the vigour of the plants
- A small computer for basic computational functions and connected to a communication module
- A cloud-based web application to process images or create 3D maps
- Captures the leaf-to-fruit ratio at a rate of 1ha/h
- Can autonomously monitor 168 hectares three times a year
- Is capable of climbing slopes up to 45º
- Is powered by and electric battery and works 8h per day
Wine is more than a beverage, is part of a culture that began to originate in the Neolithic. It was during the time of the Egyptian, Greek and Roman civilizations when wine acquired an outstanding importance, even to worship the God of the vineyards (Dionysus or Bacchus). Christianity greatly contributed to its expansion, being monasteries major centre of wine production during the Middle Ages. Currently, only in the European Union there are 1.6 million vineyards which produce about 175 million hectolitres annually.
To satisfy all kind of winegrowers and suit any palate, VINBOT arises. It is an all-terrain autonomous mobile robot equipped with a set of sensors capable of capturing and analysing vineyard images and 3D data by means of cloud computing applications. This Precision Viticulture tool will allow winegrowers estimate the yield in an accurate way and, therefore, manage the canopy optimally and harvest areas according to the ripeness of the grape. This will lead to a high control of wines’ quantity and quality.
Thus, VINBOT will help winegrowers perform three essential yield management techniques:
- Fruit removal
- Sequential harvesting by grapes ripeness by estimating the yield three times a year (spring, summer and véraison)
VINBOT will assess vineyards features by means of machine vision monitoring of main parameters hence allowing winegrowers take any needed corrective action.
VINBOT will be able to assess:
- Shoot-tip stress
- Excessive water supply after flowering
- Moderate/severe water stress
- Heat stress
- Canopy colour
- Canopy density
- Berry and cluster shape and size
- Diseases and nutrient deficiencies
All these features are related to parameters as tendrils, colour, leaf angle or gaps which will be monitored by machine vision technology integrated into VINBOT.
Despite VINBOT’s technological complexity which integrates robotics, computer vision and cloud-based applications, it will provide growers with easy-to-understand online vigour and yield maps of their vineyards, which will allow optimal management and a significant improvement of their wines’ quality result of a process from grapes with the same ripeness state.
Winegrowers will improve the organisation of the production and marketing of their wines by a better choice of the grapes to blend for each targeted wine quality segment which will increase their revenues.
The robot will have skid-steering kinematics based on four high power motor wheels. Each wheel will integrate a hub brushless motor with gearbox and encoder. The odometry will be computed with the use of the four encoders and a high precision angular sensor mounted inside the chassis.
The strong mechanical structure will allow carrying heavy loads. There are several suspension shocks possibilities. They can also be mounted at several positions to modify the robot clearance.
The robot base will navigate autonomously or remotely operated by means of a PTZ camera that will transmit video in real time.
The control architecture will be open-source and modular based on ROS.
ROS framework defines a well-organised robot software architecture and includes hundreds of user contributed packages and sets of packages called stacks, that implement functionality as localization and mapping, planning, manipulation, perception, etc.
This characteristic simplifies the software development cycle and allows easy integration and reutilization of software components whether they are device drivers or state of the art algorithms in vision, SLAM, point cloud processing, grasping, planning, etc.