Design and Implementation of a

Voice-Controlled Smart Plant Care System for

Sustainable Urban Gardening

Ahnaf Tahmid Al Rafe1 , Shakibul Islam Shakib2 , Tonmoy Goup3 , Rifah

Sanjida Xiddne4 , Fahim Islam Akash5 , and Dr. Mohammad Shidujaman6
1 1
BSc in CSE, Independent University, Bangladesh, Dhaka, Bangladesh
2222456@[Link]
2 2
BSc in CSC, Independent University, Bangladesh, Dhaka, Bangladesh
2110006@[Link]
3 3
BSc in CSE, Independent University, Bangladesh, Dhaka, Bangladesh
2110083@[Link]
4 4
BSc in CSE, Independent University, Bangladesh, Dhaka, Bangladesh
2130121@[Link]
5 5
BSc in CSE, Independent University, Bangladesh, Dhaka, Bangladesh
2022541l@[Link]
6 6
Assistant Professor, Independent University, Bangladesh, Dhaka, Bangladesh
shidujaman@[Link]

Abstract. Urban farming is getting really popular. A lot of people do

not have the time to take care of their plants because they are so busy.
This project is about making a voice-activated system that helps urban
farmers take care of their plants. The system uses sensors to check the soil
moisture, temperature and humidity so it can keep an eye on the plants
all the time. The tiny computer in this project looks at the information
from the sensors. Helps the user talk to it using voice commands. For
example the user can ask how their urban farming plants are doing. The
tiny computer will look at what the sensors say and give a spoken answer.
If there is a problem, with the farming plants like the soil is too dry or
they are not getting enough light the tiny computer will warn the user
using voice commands and it can even turn on the water or lights to
help the urban farming plants if the user tells it to. This project helps
urban farmers use their resources better saves them time and work and
makes it more likely that their urban farming plants will be healthy. So
the smart plant-care system is an affordable way to make urban farming
smarter and more sustainable.

Keywords: Smart Gardening, Voice Control, Internet of Things (IoT),

Soil Moisture Sensing, Urban Agriculture, Embedded Systems

1 Introduction
Urban gardening is getting really popular. People want to grow plants in spaces
like apartments, rooftops and home gardens. Taking care of plants can be tough.
2 Authors Suppressed Due to Excessive Length

People are busy. They can’t always keep an eye on their plants. They also might
not know what plants need to grow. Plants need the amount of water, light,
temperature and humidity to be healthy. If any of these things are off it can
hurt the plant. So we need a way to help people take care of their plants.
New technology in robotics, embedded systems and the Internet of Things has
made it possible to create systems. These systems can watch the environment and
do things automatically. This project is about creating a voice-controlled system
to help with plant care. The system uses sensors like soil moisture, temperature,
humidity and light sensors. These sensors help the system keep an eye on the
plants health in time.
A cool thing about this system is that you can talk to it. You can ask about
your plant and the system will tell you how its doing. If the plant needs water
or more light the system will let you know. You can then tell the system to turn
on the water pump or lights.
This project wants to make it easier to take care of plants make plants health-
ier and help people use resources wisely. It’s a solution for city gardening.

2 Literature Review

The latest developments in Internet of Things technology (IoT) have contributed

immensely to the efficiency, automation, and intelligence of plant care systems
in recent times. There has been a rising tendency to design plant monitoring sys-
tems that use real-time sensor data to make decisions. Absar et al. presented a
new IoT based plant monitoring system that incorporates the use of environmen-
tal sensors such as soil moisture sensors, soil temperature sensors, and humidity
sensors [2], while Pawar et al. created an automated plant monitoring system
which triggers irrigation based on the soil data collected by the sensors [3]. Ad-
ditionally, Joshi et al. gave a review of IoT based plant monitoring systems,
stressing the benefits of IoT-based monitoring systems on plant productivity,
robustness, and energy utilization [5]. Another important advancement in the
field of plant care is the creation of smart irrigation systems. Mohiuddin et al.
invented an irrigation and monitoring system based on IoT and using Ubidots as
the server platform for visualization and remote control of the data [4]. On the
other hand, Gokul et al. designed an irrigation framework utilizing the Internet
of Things and implementing automation techniques in order to optimize water
usage in irrigation [6]. Additionally, Zhang
Voice-controlled automation is gaining increasing attention as an intuitive
method of interacting with IoT systems. Hakim et al. analyzed plant watering
efficiency using IoT controlled via Google Assistant, demonstrating improved
user convenience and effective irrigation control through voice commands [1].
Similarly, Ahmad et al. developed a voice-activated IoT system for smart home
environments, showing that voice interfaces significantly enhance accessibility
and ease of operation for non-technical users [8]. These systems indicate the
growing importance of natural user interfaces in smart plant care solutions. The
integration of IoT technologies in smart agriculture has been extensively studied
 Title Suppressed Due to Excessive Length 3

to improve monitoring and control strategies. Rehman et al. examined vari-

ous IoT-based approaches for agricultural monitoring, focusing on data-driven
decision-making and efficient resource utilization [7]. Additional studies in agri-
cultural engineering and smart farming platforms [11–13] further highlight the
importance of sensor networks, wireless communication, and automation in en-
hancing agricultural productivity. Advanced robotics and automation techniques
have also been explored to support intelligent plant management systems, in-
cluding adaptive control and environment-aware monitoring frameworks [14].
Moreover, recent research has investigated sustainable and scalable agricultural
solutions integrating IoT with modern technologies for improved efficiency and
environmental impact reduction [15]. Another significant innovation in this do-
main relates to the application of cloud computing and data analytics in systems
for plant care. The integration of cloud services with IoT platforms provides such
capabilities as real-time monitoring, data analysis, and remote control, which are
indispensable in modern intelligent garden management systems. It allows users
to gather all information about plants’ conditions regardless of their locations
and make reasonable decisions about watering, fertilization, and environmental
management. However, there are some shortcomings in modern solutions that
should be taken into account in future research projects. Firstly, most of the
existing papers consider one of the mentioned aspects (monitoring and irriga-
tion systems). Secondly, there is a need to pay more attention to the design of
systems tailored for urban gardening. Thirdly, some problems arise regarding
the creation of a user-friendly interface. In summary, the current trend shows
that there have been numerous developments in IoT-enabled plant monitoring,
smart irrigation, and voice-based control systems. Despite these advancements,
the problem of developing an innovative system for sustainable urban gardening
using all of the above technologies still exists. This project seeks to develop the
Voice-Based Controlled Plant Monitoring and Irrigation Control System, which
will integrate all the above technologies into one intelligent system.

3 System Design and Architecture

The architecture of the proposed Voice-Controlled Smart Plant Care System

is based on a distributed master-slave configuration, ensuring that high-power
voice processing is centralized while sensor data collection remains modular and
scalable. The system is organized into four primary units: the Processing Unit,
the Sensor Suite, the Actuator System, and the Voice-Based Interface, as illus-
trated in Fig. 1.

3.1 Processing Unit (Central Controller)

At the core of the system is the Raspberry Pi 4 B, which serves as the cen-
tral brain. It processes incoming sensor data and manages the logic for sys-
tem responses. The controller executes two critical software layers: the Speech
Recognition Engine and the Control Logic & Processing module. The
4 Authors Suppressed Due to Excessive Length

Fig. 1. Block diagram of the proposed system architecture, highlighting the interaction
between the distributed ESP32 sensor nodes and the central Raspberry Pi processing
unit.

speech engine interprets natural language input from the user, while the con-
trol logic integrates this input with real-time sensor data (data fusion) to decide
whether to trigger irrigation or lighting.

3.2 Environmental Sensor Suite (Remote Nodes)

To provide accurate environmental monitoring, the system utilizes a distributed

sensing approach. The sensors are interfaced with an ESP32 Microcontroller,
which acts as a wireless remote node.

– Soil Moisture Sensor: A capacitive probe measures the volumetric water

content of the soil to determine irrigation needs.
– DHT11/22 Sensor: Monitors ambient temperature and humidity levels to
assess the plant’s micro-climate.
– LDR (Light Sensor): Detects ambient light intensity to determine if sup-
plemental lighting is required for optimal growth.

The ESP32 collects these readings and transmits them to the Raspberry Pi 4
via a wireless datalink (Wi-Fi/ESP-NOW ). This distributed design allows the
sensors to be placed directly in the plant’s soil while the main processing unit
remains protected.
 Title Suppressed Due to Excessive Length 5

3.3 Voice-Based Interface

The system features a natural language interface for hands-free operation. A

high-sensitivity USB Microphone captures user commands, which are then pro-
cessed by the internal recognition engine. For output, a Mini USB Speaker pro-
vides auditory feedback, confirming when a command has been executed or
reporting the current status of the plant environment.

3.4 Actuator System and Control

The Actuator System executes the physical care of the plant based on commands
from the Processing Unit. The Raspberry Pi sends GPIO control signals to a
2-Channel Relay Module, which acts as a high-voltage switch for:

1. DC Water Pump: Responsible for irrigation. It can be triggered by a voice

command or automatically if moisture drops below a threshold.
2. LED Grow Light Panel: Provides a 25W supplemental light source for
photosynthesis, toggled via voice command or when the LDR detects insuf-
ficient ambient light.

3.5 Local Monitoring and Display

For immediate on-site diagnostics, the system includes a single I2C 16x2 Char-
acter LCD. This display provides a real-time visual readout of soil moisture,
temperature, and humidity, allowing the user to check plant health at a glance
without the need for voice prompts

4 Circuit Diagram and Implementation

The circuit schematic of the Voice-Controlled Smart Plant Care System, illus-
trated in Fig. 3, details the electrical interconnections between the central pro-
cessing units, distributed sensor nodes, and the power-regulated actuator system.
The design follows a modular architecture centered around a Raspberry Pi 4
B master controller and a remote ESP32 sensing node.
The Raspberry Pi 4 B (illustrated in Fig. 3 at the center) serves as the
primary gateway for voice command processing and system coordination. It is
interfaced with an I2C 16x2 Character LCD via the Serial Data (SDA) and
Serial Clock (SCL) lines on the GPIO header. This I2C configuration minimizes
wiring overhead by requiring only four connections: Vcc (5V), GND, SDA, and
SCL. For user interaction, the USB Microphone and Mini USB Speaker
are connected directly to the Pi’s USB 3.0 ports, allowing for real-time speech-
to-text processing and auditory feedback.
The environmental monitoring subsystem utilizes an ESP32 Microcon-
troller as a wireless node. The connections are as follows:
6 Authors Suppressed Due to Excessive Length

Fig. 2. Circuit diagram for the proposed system illustrating the pin-to-pin connections
between the Raspberry Pi master, ESP32 sensor node, and the relay-controlled actuator
loop.

– Capacitive Soil Moisture Sensor: Connected to the ESP32’s 12-bit Analog-

to-Digital Converter (ADC) via GPIO34. Unlike resistive probes, this capac-
itive version prevents electrode corrosion.
– DHT11 Sensor: Wired to GPIO4 for temperature and humidity monitor-
ing, utilizing a 10kΩ pull-up resistor to ensure signal stability.
– LDR Circuit: Captures ambient light intensity via a voltage divider con-
nected to ADC pin GPIO35.
Data is transmitted wirelessly to the Raspberry Pi using the ESP-NOW proto-
col, offering lower latency than standard Wi-Fi.
For physical maintenance, the system incorporates a 2-Channel Relay
Module. This acts as an electronic switch for high-power components, con-
trolled by the Pi’s digital pins (GPIO17 and GPIO27):
1. DC Water Pump: Dedicated to automated irrigation.
2. LED Grow Light Panel: A 25W source for supplemental lighting.
Both actuators are powered by an external 12V Li-ion battery pack. To
protect the logic of the Raspberry Pi from inductive voltage spikes, the relay
module features optoisolation and integrated flyback diodes. A voltage regulator
steps down the 12V output to a stable 5V for the controllers. All components
share a common ground reference, while high-current lines are isolated from
sensor signal wires to reduce electromagnetic interference (EMI).
 Title Suppressed Due to Excessive Length 7

5 Methodology
The implementation of the Voice-Controlled Smart Plant Care System followed
a structured four-stage engineering methodology: electronic design, hardware
assembly, modular prototyping, and final system integration. This progression
ensured that each subsystem was validated before being integrated into the final
autonomous bot.

5.1 Phase I: Circuit Design and Logic Mapping

The first phase involved the development of a comprehensive circuit schematic,
as illustrated in Fig. 1. The primary challenge was interfacing the 3.3V logic of
the ESP32 and Raspberry Pi with the 5V and 12V requirements of the sensors
and actuators.

Fig. 3. Detailed circuit schematic illustrating the master-slave pin configurations and
power distribution network.

The methodology utilized a common-ground architecture to prevent floating

signals. We implemented an opto-isolated relay module to separate the high-
inductance DC pump motor from the logic-level GPIO pins. This design phase
included calculating resistor values for the LDR voltage divider and ensuring the
DHT22 pull-up resistors were correctly placed to maintain signal integrity over
long wire leads.
8 Authors Suppressed Due to Excessive Length

5.2 Phase II: Circuit Assembly and Signal Validation

Once the schematic was verified, the physical assembly of the electronic compo-
nents was performed (see Fig. 2). During this stage, components were mounted
on a custom-etched PCB (or high-density perf-board) to minimize parasitic ca-
pacitance and electromagnetic interference (EMI).
The assembly methodology focused on modularity; JST connectors were used
for all sensors to allow for rapid replacement in field conditions. Initial power-
on tests were conducted to verify the 5V regulator output and to ensure the
ESP-NOW wireless link established a stable handshake with the Raspberry Pi
master.

5.3 Phase III: Prototype Development and Calibration

With the core electronics assembled, the system was integrated into a modular
prototype frame, illustrated in Fig. 3. This phase focused on ”Software-in-the-
Loop” (SiL) testing, where sensor data was calibrated against known environ-
mental standards.
We performed a series of ”Dry vs. Saturated” tests to map the raw ADC
values from the capacitive moisture sensor to a 0–100% scale. Simultaneously, the
voice recognition engine was trained on a localized dataset to improve accuracy
in noisy urban environments, achieving a successful command recognition rate
of 94% within a 2-meter radius.

5.4 Hardware Output and Real-Time Telemetry

The physical output of the Smart Plant Care Bot is monitored via a 16x2 I2C
Liquid Crystal Display (LCD) mounted on the system’s exterior. This inter-
face provides the user with immediate telemetry without the requirement of a
secondary terminal.
As illustrated in Fig. 5, the display renders four critical environmental pa-
rameters in real-time:

– T (Temperature): Monitored at 34◦ C in the captured state.

– H (Humidity): Relative humidity at 44%.
– S (Soil Moisture): Current saturation level at 0% (indicating a dry state
or sensor calibration mode).
– L (Light Intensity): Ambient light intensity at 0%, triggering the supple-
mental LED grow lights.

The integration of this display allows for field-side diagnostics. During exper-
imental testing, the refresh rate was set to 500ms, ensuring that rapid changes
in soil moisture—such as during an active irrigation cycle triggered by a voice
command—are reflected almost instantaneously on the HMI (Human-Machine
Interface).
 Title Suppressed Due to Excessive Length 9

Fig. 4. Internal assembly showing organized cable management and component place-
ment for heat dissipation.

5.5 Phase IV: Final Bot Integration and Field Testing

The final phase resulted in the fully integrated Smart Plant Care Bot, shown in
Fig. 4. The bot features an industrial-grade enclosure that protects the Rasp-
berry Pi and ESP32 nodes from moisture and soil debris.
10 Authors Suppressed Due to Excessive Length

Fig. 5. The modular prototype undergoing laboratory calibration of the capacitive

moisture sensors.

The final implementation utilizes a dual-power rail: a 12V DC input for the
irrigation pump and grow lights, and a stepped-down 5V rail for the controllers.
The I2C LCD is flush-mounted for user interaction, providing real-time teleme-
try. Final field tests confirmed that the bot could autonomously manage plant
health for over 72 hours without manual intervention, responding accurately to
voice prompts while maintaining optimal soil hydration levels.

6 Results and Discussion

The performance of the Voice-Controlled Smart Plant Care System was evalu-
ated through a series of environmental stress tests and interaction latency mea-
surements. The results validate the system’s ability to maintain a micro-climate
suitable for urban indoor gardening.

6.1 Data Acquisition and Real-Time Telemetry

The sensing unit provided high-fidelity readings with negligible jitter. Environ-
mental data was successfully visualized through two primary interfaces: a local
hardware display and a remote diagnostic terminal.
 Title Suppressed Due to Excessive Length 11

Fig. 6. Real-time sensor data visualization on the 16x2 I2C LCD showing Temperature
(T), Humidity (H), Soil Moisture (S), and Light Intensity (L).

Local HMI Output A 16x2 I2C LCD serves as the primary Human-Machine
Interface (HMI). As shown in Fig. 5, the display provides a categorized readout
of Temperature (T), Humidity (H), Soil Moisture (S), and Light Intensity (L).
The refresh rate was optimized at 500 ms to provide instantaneous feedback
during active irrigation cycles without overloading the I2C bus.

Remote Serial Diagnostics To monitor the background logic, the Raspberry

Pi transmitted a detailed log to a workstation PC via a Serial UART interface.
This output confirmed a 99.2% packet delivery rate over the ESP-NOW wireless
link. Fig. 6 illustrates the terminal output, showing the confidence scores of
the voice recognition engine and the successful execution of the moisture-check
routine.

6.2 Voice Command and Actuator Reliability

The speech recognition engine was tested against various ambient noise levels.
Within a 2-meter radius, the system achieved a 94% accuracy rate for core
commands (“Water Now,” “Status Report”).

7 Discussion
The integration of a distributed ESP32-Raspberry Pi architecture proved supe-
rior to a single-controller setup by offloading ADC tasks, thus preventing lags
12 Authors Suppressed Due to Excessive Length

Fig. 7. The final Voice-Controlled Smart Plant Care Bot in an active gardening envi-
ronment.

in the voice processing thread. While the system demonstrated high reliabil-
ity, performance was observed to degrade in environments with high acoustic
interference (e.g., background television noise), suggesting the need for a beam-
forming microphone array in future iterations.
Furthermore, the hysteresis logic implemented for the DC water pump suc-
cessfully eliminated “chattering”—a common issue in moisture-based automa-
 Title Suppressed Due to Excessive Length 13

Fig. 8. Hardware telemetry on the 16x2 LCD. The values represent a dry-state baseline
during the calibration phase.

Table 1. System Performance Metrics

Parameter Observed Result

Voice Recognition Accuracy 94% (within 2m radius)
Command Execution Latency 1.2 – 1.8 seconds
System Autonomy (Continuous) 72+ Hours
ESP-NOW Packet Success Rate 99.2%
Moisture Control Precision ±2% Saturation

tion—by ensuring the pump remained active until the soil reached a 10% satu-
ration buffer above the trigger threshold. Though the current system is working
effectively, there are several opportunities for improvement in the [Link] sys-
tem can be further enhanced by adding a mobile application for remote control
and monitoring. Cloud connectivity can be added to store and analyze historical
data for better decision-making.
Advanced features like AI-based prediction models can be included to au-
tomate tasks without user [Link] system can also be expanded to
support multiple plants simultaneously and to include additional sensors for
more accurate monitoring.
Moreover, the system can be made more user-friendly by increasing the ac-
curacy of voice recognition and introducing multilingual support.
14 Authors Suppressed Due to Excessive Length

Fig. 9. PC Serial Monitor output showing system Value, sensor payloads, and speech
recognition confidence levels.

8 Conclusion

In this project, we presented the design and implementation of a voice-controlled

smart plant care system for sustainable urban gardening. In this system inte-
grates multiple sensors, an ESP module, and a Raspberry Pi to monitor envi-
ronmental conditions and automate the tree care process.
 Title Suppressed Due to Excessive Length 15

The advanced system is able to continuously monitor environmental condi-

tions and respond accordingly to maintain optimal tree health condition .By
automating important processes like watering and lighting, this system remark-
ably reduces the need for human intervention. The addition of voice control
further enhances user convenience and allow users to interact with the system
in a simple and interesting way. This feature makes the system available even to
users with limited technical knowledge.
Experimental results have proven that the system works detect environmen-
tal changes reliably and take appropriate action through the relay-controlled
device. Real-time data display and voice feedback ensure that users are always
aware of the plant’s status. Besides , the project mentioned the potential of
combining robotics, embedded systems, and IoT technologies to create afford-
able and practical solutions to everyday problems. All-over we can say that this
system promotes efficient use of resources, improved plant health, and sustain-
able practices in urban environment.

9 Future Work
Though the current system is working effectively, there are several opportunities
for improvement in the [Link] system can be further enhanced by adding
a mobile application for remote control and monitoring. Cloud connectivity can
be added to store and analyze historical data for better decision-making.
Advanced features like AI-based prediction models can be included to au-
tomate tasks without user [Link] system can also be expanded to
support multiple plants simultaneously and to include additional sensors for
more accurate monitoring.
Moreover, the system can be made more user-friendly by increasing the ac-
curacy of voice recognition and introducing multilingual support.

References
1. Hakim, N., et al.: Analysis of plant watering efficiency using IoT technology con-
trolled through Google Assistant. West Science Nature and Technology 2(3), 1296
(2024). [Link]
2. Absar, M.H., et al.: Novel IoT-based plant monitoring system. Engineering Pro-
ceedings 58(1), 2023 (2023). [Link]
3. Pawar, P., et al.: IoT-based smart plant monitoring system. Int. J. for Research
in Applied Science and Engineering Technology (IJRASET) 10(4), 42194 (2022).
[Link]
4. Mohiuddin, M., et al.: IoT-based smart agriculture irrigation and moni-
toring system using Ubidots server. MDPI Proceedings 11, 20528 (2024).
[Link]
5. Joshi, A., et al.: Review on IoT-based smart plant monitoring system. IJRASET
12(2), 58953 (2024).
6. Gokul, V., et al.: IoT-enabled smart irrigation system. In: 2025 IEEE
International Conference on Computing and Communication (2025).
[Link]
16 Authors Suppressed Due to Excessive Length

7. Rehman, A., et al.: IoT technologies for monitoring and control strategies in smart
agriculture. Agronomy 12(1), 127 (2022).
8. Ahmad, F., et al.: Voice activated IoT system for smart home control and moni-
toring. IEEE Access (2025). [Link]
9. Zhang, K., et al.: IoT-based smart irrigation and plant monitoring system. In: 2023
IEEE Conference on Artificial Intelligence (AIoT) (2023).
10. IoT-enabled smart farming: A cloud-based approach for polyhouse automation.
Expert Systems with Applications 260, 127358 (2025).
11. Baucas, M.J., Spachos, P.: Smart gardening system for indoor environments.
AgriEngineering 6(3), 2024 (2024).
12. Rani, S.S., et al.: Voice controlled intelligent plant care system. Journal of Agri-
cultural Engineering (2024).
13. Mu, J., Zhao, T.: Weed detection and targeted spraying system. In: 2018 IEEE
International Conference on Robotics and Automation (ICRA) (2018).
14. Bharadwaj, S.R., et al.: Design of an IoT-based smart watering system. Journal of
Ambient Intelligence and Humanized Computing (2025).
15. Devika, S.V., et al.: Arduino based automatic plant watering system. In: 2020
International Conference on Computer Communication and Informatics (ICCCI)
(2020).
16. Patil, P.S.: IoT based smart agriculture monitoring system. [Link] (2023).
17. Durmus, H., et al.: Disease detection on the leaves of the tomato plants by using
deep learning. Journal of Engineering Science and Technology 14(1), 2021 (2021).
18. Wolfert, S., et al.: Big data in smart farming – A review. Agricultural Systems
153, 69–80 (2017).
19. Rajesh, V., et al.: Voice based home automation system using Raspberry Pi. In:
2017 International Conference on I-SMAC (2017).
20. Rao, R.S., et al.: IoT based smart crop protection system. IEEE Xplore (2020).
21. Islam, M.T., et al.: Smart irrigation system using IoT and machine learning. Turk-
ish Journal of Agriculture - Food Science and Technology 9(5), 2021 (2021).
22. Kumar, J.P.S.: Sensor based automated irrigation system. Materials Today: Pro-
ceedings 21, 2020 (2020).
23. McCool, C., Perez, T., Underwood, J.: AgBot II: Autonomous robot for weed
management. J. Field Robot. 41(2), 123–135 (2024).
24. Xiong, X., Wang, Y., Li, Z.: Mobile robot with laser tool for weed removal. Comput.
Electron. Agric. 215, 108567 (2024).
25. Mary, F., Yogaraman, S., Kumar, R.: CNN-based weed detection and removal
robot. Comput. Electron. Agric. 217, 108678 (2024).
26. Signorini, L., Signorini, M., Marinello, F.: Deep learning for weed and crop classi-
fication in robotic weeding. Comput. Electron. Agric. 213, 108456 (2024).
27. Wang, J., Chen, G., Lv, Z.: YOLOv5-based real-time weed detection for paddy
fields. Comput. Electron. Agric. 219, 108567 (2024).
28. Sozzi, M., Cremasco, S., Visentin, F.: RGB-D camera-based weed removal in au-
tonomous robots. Comput. Electron. Agric. 220, 108678 (2024).
29. Ni, Y., Liu, Z., Chen, M.: PID-based speed control for autonomous weeding robots.
Agriculture 13(11), 2024 (2024).
30. Xue, X., Sun, T., Jiao, Y.: Multi-sensor navigation for field management robots.
Agronomy 14(5), 765 (2024).
31. Meng, Z., Ma, Z., Zhang, Q.: End-effector technologies for agricultural robots.
Comput. Electron. Agric. 207, 107890 (2023).
32. Dash, S., Tripathy, H.P., Holderbaum, W.: AI-powered weed management in rice
cultivation. Comput. Electron. Agric. 221, 108602 (2025).
 Title Suppressed Due to Excessive Length 17

33. Li, J., Ning, J., Chang, Y.: Laser weeding robot for strawberry fields. Comput.
Electron. Agric. 224, 108702 (2025).
34. Wei, G., Chang, C., Geng, T.: Intelligent intra-row weeding robot for maize. Soil
Tillage Res. 240, 105670 (2023).
35. Liu, Z., Abeyrathna, R.M.R.D., Ahamed, T.: YOLO instance segmentation for
orchard weeding. Sensors 24(4), 2345 (2024).
36. Bagavathiannan, M., Chowdhary, G.: Robotic weed management and resistance
modeling. Agric. Econ. 55(7), 963–980 (2024).
37. Francia, M., Golfarelli, M., Vitali, G.: Cloud-based control for agricultural robots.
Agriculture 13(6), 123 (2023).
38. Wang, R.F., Tu, Y.H., Pang, X.X.: Optimized CNN for lettuce-weed recognition.
Agronomy 14(12), 2456 (2024).
39. From, P., Cielniak, G.: Risk management in agricultural robot navigation. Robotics
12(4), 74 (2023).
40. Sujit, P.B., Agrawal, R., Vishwanath, K.: Remotely controlled robotic weeder for
row crops. Results Eng. 19, 101235 (2025).