Solar Powered Automated Cutting Machine

Project Title

Solar Powered Automated Cutting Machine

Project Area of Specialization RoboticsProject Summary

The solar powered grass cutting machine is an autonomous robot that cuts the grass using Artificial intelligence based machine learning and deep learning algorithms. It detects boundaries of the grean area with color sensors and camera images that are processed using color coding algorithms. The primary system of the robot is based on three motors and two sensors. Initial two motors are responsible for the movement of robot and the third motor with embedded controller unit is used for the cutting of grass. The reverse movement of robot is also controlled by IR sensors. There wil lbe a camera on top of the robot that acts like an human eye. The camera can be rotated 360 degrees and all view would be visible. There are two modes of operation. First mode is fully automated that manages the machine to be operated at its own. In the second mode, semi supervised, the machine will be controlled through a wifi controller that has a range of 500 meters. The range is extendable and with the addition of more routers, it can be extended to 1000 meters. In the semisupervised mode, the machine can be controlled using a laptop or an android app from inside the house. The process is similar to flying a drone where the operator is far away from the plane and using GPS and microcontrollers, the drone can be controlled. In the same way, the operator of the robotic grass cutting machine will be sitting inside the air conditioned house and using a laptop, the machine (placed outside in the lawn) can be controlled using a rotating camera that will provide complete 360 degree view of the scene.  The complete system is attached to a microcontroller that functions like a human brain. Another key feature of the robot is high strength solar batteries. As grass cutting requires enormous power, so a hybrid  engine is required that is fully functioned on the chargeable solar batteries. After being switched on, robot will identify the grass area and will do initial testing for the length of the grass and the amount of weed and presence of other plants in the target area. When initial tests are passed, a starting point is being selected and the cutting area is estimated. Based on the length of the grass, cutting power is adjusted and using IR sensors, robot starts its cutting tasks and avoids any obstacles in its path. The second motor controls the height of the blades. Acording to the lenght of the grass, the height of the blades is adjusted and keeps on varying with variation in the grass length. The machine is solar opertaed. Its all three motors are powered by solar batteries. As it is an outdoor machine, so there would be plenty of sunshine and motors will be charged all the time. Also using autonomus operation, cutting of one or two canals of grass will not take more than 10 minutes. The strength of the batteries are such that four hours of continuous operation is possible with a single recharge and it will take only half an hour to fully recharge the batteries.

Project Objectives

Following are the main objectives for the proposed research project.

• In automated mode, fully automatic operation is performed without any human intervention.
• In the controlled mode, the machine is operated using a laptop or even a mobile phone.
• A novel computer vision project that identifies the green area of grass.
• Solar powered Robot cutting machine (SRCM) can also distinguish between grass and other crops such as weed etc.
• On completion of pilot and prototype project, a commercial version will be built keeping in view the economic aspects of the project.

Project Implementation Method

Deep Learning techniques will be the core program software in the proposed project. The primary function for the robotic cutting machine is to detect cut-able grass and identify between grass and weeds. To achieve his goal, convolutional neural network algorithms will be employed. Initially the system will be trained on various types and areas of grass along-with different kinds of weeds. The proposed system will also be trained on the required lengths of grass to be cut. To detect the obstacles such as trees and plants, infrared sensors will be used and on detecting any obstacle on its path, robotic machine will change its direction and after crossing the obstacle, it will return to its original path.

Benefits of the Project

In this world of innovations there is a dire need for fast and efficient work. Our project reduces man power, time and increases work efficiency. It is based on a robotic action that automatically cuts the grass using solar powered batteries. Thus it saves power and time and even grass is cut smoothly and efficiently in a fraction of time (at least less than 5 times if the same area is cut by humans).

It is a commercial product and masses will be benefited in general. Especially government agencies where it hires a large number of manpower to maintain lawns, sport centers and the environment. Using solar powerede robotic grass cutting machine, the burden to pay a large workforce is reduced and the natural environment is maintained in an efficient manner.

Technical Details of Final Deliverable

Final deliverable will be an artificial intelligent solar powered robotic grass cutting machine that uses deep learning algorithms to identify grass and differentiate between grass and weeds. For the movement of robot, Rasberry pi microcontroller is used. It is based on three motors that controls front and backward movement of the mower. The height of the blades is adjusted using an auxiliary motor that has a vertical up and down movement. The camers detect the length of the grass and using perpendicular intercept algorithm, the up and down movement of the blades is controlled. Thus an efficient cutting of the grass with fine details is achieved that is much faster (more than 5 times) and pleasant looking than human efforts. 

Final Deliverable of the Project HW/SW integrated systemType of Industry Agriculture Technologies Artificial Intelligence(AI), RoboticsSustainable Development Goals Industry, Innovation and Infrastructure, Life on LandRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Induction Motors	Equipment	3	10000	30000
Rasberry Pi Micro Controller	Equipment	1	8000	8000
IR Sensors	Equipment	3	6000	18000
HD Camera	Equipment	1	14000	14000
Solar plates	Miscellaneous	2	3000	6000
Solar batteries	Miscellaneous	1	4000	4000