Weed identification and control using autonomous mobile robot
The purpose of this work is to develop an autonomous robot for weed control in fields. Non-chemical weed control is an efficient method to remove weeds in field and that can reduce the dependency of farmer's efforts. Usually weeds are removed by manual technique which can increase the
2025-06-28 16:36:46 - Adil Khan
Weed identification and control using autonomous mobile robot
Project Area of Specialization RoboticsProject SummaryThe purpose of this work is to develop an autonomous robot for weed control in fields. Non-chemical weed control is an efficient method to remove weeds in field and that can reduce the dependency of farmer's efforts. Usually weeds are removed by manual technique which can increase the labor cost. In past, the robot required a large amount of power supply to produce electric spark of 15 KV to kill weeds whereas in this project weeds are removed by hoe tool which is mounted at the bottom end of robotic arm to kill weeds by pinching them down. The proposed weed control robot comprises of a moving platform mounted with a delta robotic arm which has three degrees of freedom (DOFs) to access the weeds anywhere in the field. The delta manipulator has high acceleration due to its light weight. It can operate in natural and complex environment where working conditions are changed time to time. It employs two vision systems: one is primary vision system and other is secondary vision system. Primary vision system mounted at the front side of moving vehicle to detect, acquire and analyze the weeds and send their coordinates to robotic arm. While secondary vision system relocates the weeds individually that detected by primary vision system to increase the accuracy of weeds detection. Weeds are remove by hoe tool which pinch down and cutoff the connections of weeds from the soil.
Project ObjectivesThe objectives of this project are:
1. To develop a 3-DOF delta robot for weed removal.
2. Development of vision based position control system for the delta robotic arm.
3. Development of a mobile robot platform that carries the delta robot and image processing unit in field where weed control is required.
4. Develop machine vision system for weed detection and classification.
Project Implementation MethodThe moving vehicle will move autonomously in the crop field, the primary camera which is mounted at front side of the moving vehicle first detect weeds among the crops by making a comparison between the current and previously saved image and send its coordinates to the secondary camera which is mounted at the bottom edge of the Hoe tool. The secondary camera will take a new image of the weed and compare it with the coordinates of weed obtained from the previous image taken by the primary camera and verified it either it is a weed or a crop by machine vision algorithm. After the identification of weed the secondary camera will send its position coordinates to the delta manipulator controller developed in a raspberry pi computer. The controller will then command and control the servo motors to adjust the angles of all the three limbs attached with the shafts of the motors to position the end-effector at the specified position of the weed and perform the pinching process up to specified number of times. The ‘Hoe tool’ which is mounted at with the end-effector of robotic arm will dig-out the detected weed.
Benefits of the ProjectBenifits of the proposed weed control robot are:
-
Labor cast will be decreased.
-
No herbicide usage will be involved.
-
The developed mobile robot will works autonomously and no manual interference will be required.
-
Renewable solar energy can be used to power up the developed robot which will save money that means its operation requires only maintenance cast.
-
The utilization of the proposed weed control robot will effectively control the weed which in turn will increase the crop production as compared to conventional weed control methods.
-
No hazardous effects on plant growth will appear because of herbicideless weed control which will also help to avoid the atmospheric pollution.
In the final form of the project, it will contain a delta robot mounted on a mobile robot that can autonmously navigate in the filed among crop lines and will autonomously detect the weed plants among crops to eliminate them.
Final Deliverable of the Project HW/SW integrated systemCore Industry AgricultureOther IndustriesCore Technology RoboticsOther TechnologiesSustainable Development Goals Good Health and Well-Being for People, Decent Work and Economic Growth, Industry, Innovation and InfrastructureRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| Arduino DUE | Equipment | 4 | 2000 | 8000 |
| DC Geared motor with rotary encoder | Equipment | 5 | 1700 | 8500 |
| backalite sheet | Equipment | 1 | 1100 | 1100 |
| steel frame to mount the delta robot arms | Equipment | 1 | 3000 | 3000 |
| Russberry Pi 4 | Equipment | 2 | 18000 | 36000 |
| Digital Cameras | Equipment | 2 | 2500 | 5000 |
| Electronic loose components | Equipment | 1 | 2000 | 2000 |
| Delta robot End rod bearings | Equipment | 3 | 800 | 2400 |
| delta robot links | Equipment | 3 | 500 | 1500 |
| H-bridge | Equipment | 5 | 500 | 2500 |
| Labour cost of mobile robot developement | Miscellaneous | 1 | 5000 | 5000 |
| Printing | Miscellaneous | 1 | 5000 | 5000 |