Indoor Flower-Watering Robot Navigates Smart Homes with Precision
In a world where automation is reshaping daily life, a new robotic innovation from researchers in China is bringing the future of indoor gardening a step closer to reality. A team of engineers has developed a mobile, intelligent flower-watering robot capable of autonomously navigating indoor environments, identifying plants in need of water, and delivering precise irrigation—all without human intervention. This breakthrough, detailed in a recent study published in Ordnance Industry Automation, demonstrates how embedded systems, sensor fusion, and robotic mobility can converge to solve a common yet overlooked domestic challenge: keeping indoor plants alive during long absences or busy schedules.
The research, led by Lyu Pengfei from the School of Mechanical & Transportation Engineering at Guangxi University of Science & Technology, introduces a fully integrated robotic system designed specifically for indoor horticultural maintenance. Unlike traditional irrigation systems that rely on fixed piping or scheduled timers, this robot moves on a wheeled platform, follows pre-mapped paths, and makes real-time decisions based on environmental data. It represents a shift from static automation to dynamic, mobile service robotics—a trend gaining momentum in smart homes and commercial indoor spaces.
The motivation behind the project stems from both practical and societal shifts. As urban populations grow and indoor greenery becomes a staple in offices, shopping centers, and public buildings, the demand for low-maintenance plant care is rising. At the same time, an aging population and shrinking workforce in many countries are increasing the need for labor-saving technologies. Manual plant watering, while seemingly simple, consumes time and is prone to inconsistency—especially in large facilities where hundreds of potted plants may be scattered across vast floor plans. The team recognized that a mobile robotic solution could not only reduce labor costs but also improve watering accuracy, prevent over- or under-watering, and ultimately extend the life of indoor vegetation.
At the heart of the robot is a sophisticated control system built around the STM32F4, a 32-bit ARM Cortex-M4 microcontroller known for its high processing speed, low power consumption, and rich peripheral support. This central processing unit acts as the robot’s brain, coordinating inputs from multiple sensors and executing navigation, decision-making, and actuation commands. The choice of an embedded system reflects a growing trend in robotics: leveraging compact, energy-efficient computing platforms that can run complex algorithms without the need for external servers or cloud connectivity—ensuring reliability even in environments with limited network access.
One of the most innovative aspects of the design is its navigation strategy. Instead of relying on GPS—which is ineffective indoors—or complex simultaneous localization and mapping (SLAM) systems that require significant computational power, the robot uses a line-following approach guided by a linear CCD image sensor, the TSL1401. This sensor, mounted on the front of the robot, captures a 128-pixel grayscale image of the floor surface, allowing the system to detect dark lines painted or taped on the ground. By continuously analyzing the position of the line within its field of view, the robot can adjust its wheel speeds to stay on course, making sharp turns or straight runs as needed.
This method offers several advantages. First, it is highly reliable in controlled indoor environments where lighting conditions can be managed. Second, it requires minimal infrastructure—simply laying down a reflective or contrasting tape path allows the robot to operate. Third, the linear CCD provides high-resolution data compared to simpler infrared line sensors, enabling more precise path tracking and smoother motion. The researchers noted that during testing, the robot maintained a consistent trajectory with minimal oscillation, even when navigating curves or intersections.
But navigation is only half the challenge. The robot must also determine when and where to water. To address this, the team integrated a wireless soil moisture detection system. Small sensor nodes, each equipped with an MSP430 microcontroller and an MS10 soil moisture sensor, are placed in or near target planters. These nodes periodically measure the water content in the soil and transmit the data via NRF24L01 wireless modules—2.4 GHz radio transceivers known for their low power consumption and robust indoor performance—to the robot’s main control board.
When the robot begins its patrol, it queries the sensor network to identify which plants have fallen below a user-defined moisture threshold. This selective approach prevents unnecessary watering and conserves both water and battery life. The system’s wireless architecture allows for scalability; additional sensors can be added without modifying the robot’s hardware, making it suitable for deployment in large office complexes or botanical exhibits.
Once the robot reaches a target plant, it activates its watering mechanism—a three-axis RPP-type robotic arm mounted on the platform. The “RPP” designation refers to the arm’s kinematic structure: one rotational joint at the base, followed by two prismatic (linear) joints that extend and retract the arm vertically and horizontally. This configuration allows the arm to reach over tall plants like the green philodendron, which can grow up to 1.5 meters in height, and adjust its spray nozzle to an optimal position above the soil surface.
The end effector features a four-bar linkage mechanism controlled by a servo motor, enabling it to open and close like a scissor, adjusting the spray pattern to match the size of the pot. A small water pump, powered by a 24-volt lithium battery system, delivers water from a 15-kilogram onboard tank through a flexible hose to the nozzle. The entire watering sequence—approach, arm deployment, spraying, and retraction—is fully automated and takes less than 30 seconds per plant.
To ensure safety and adaptability, the robot is equipped with ultrasonic sensors (KS106 model) for obstacle detection. These sensors emit high-frequency sound waves and measure the time it takes for the echo to return, calculating the distance to any object in front of the robot. With a detection range of up to 6 meters and a refresh rate of 50 Hz, the system can identify people, furniture, or other obstacles in real time. If an obstruction is detected, the robot can pause, wait, or reroute along an alternative path if the environment allows.
The robot operates in two modes: autonomous and manual. In autonomous mode, it follows the pre-laid path, checks moisture levels, and waters plants without human input. In manual mode, an operator can control the robot remotely using a wireless joystick or smartphone interface, useful for repositioning or troubleshooting. This dual-mode functionality enhances usability, particularly during initial setup or in dynamic environments where the path may need temporary adjustment.
From a mechanical standpoint, the robot’s design prioritizes stability and compactness. The chassis is a two-tiered structure, 600 mm long, 650 mm wide, and 1,200 mm tall, weighing approximately 50 kg when fully loaded. The lower tier houses the water tank and battery, lowering the center of gravity and improving balance. The upper tier contains the control electronics and the robotic arm base. The drive system consists of two independently controlled DC motors powering the left and right wheels, with two passive caster wheels at the front and back for support. This differential drive configuration allows the robot to turn in place, a crucial feature for maneuvering in tight corridors.
The team conducted extensive kinematic modeling to understand the relationship between wheel speed and the robot’s overall motion. By analyzing the velocity of each drive wheel, they derived equations that describe the robot’s linear and angular movement, enabling precise control over its trajectory. This mathematical foundation is essential for predictable behavior, especially when integrating sensor feedback into the control loop. The results showed that the robot could achieve smooth acceleration, accurate turning, and stable straight-line travel—key requirements for reliable indoor operation.
During experimental trials, the robot demonstrated consistent performance across multiple test scenarios. It successfully navigated a simulated office layout with 10 potted plants spaced evenly along a black tape path. When soil sensors indicated low moisture, the robot stopped, extended its arm, and delivered a calibrated amount of water before continuing to the next station. The entire cycle, including return to the charging station, was completed without human intervention. Operators noted that the system was quiet, reliable, and intuitive to configure.
One of the most significant advantages of this mobile approach is its flexibility. Unlike fixed irrigation systems, which require permanent installation and can disrupt interior design, the robot can be deployed temporarily or relocated as needed. It does not require plumbing modifications, making it ideal for rental spaces, temporary exhibitions, or historical buildings where structural changes are restricted. The modular design also allows for future upgrades—such as adding cameras for plant health monitoring or integrating with smart building management systems.
The implications of this technology extend beyond convenience. In commercial settings, such robots could reduce the workload of janitorial and maintenance staff, allowing them to focus on higher-value tasks. In healthcare facilities or senior living communities, where indoor plants are used for therapeutic purposes, automated watering ensures that greenery remains vibrant even when staff are stretched thin. The system could also be adapted for use in greenhouses, vertical farms, or research laboratories where precise environmental control is critical.
Moreover, the robot contributes to sustainability. By watering only when necessary and delivering the right amount, it minimizes water waste—a growing concern in urban areas. The use of rechargeable lithium batteries further reduces environmental impact compared to disposable power sources. The researchers emphasized that their design prioritizes energy efficiency, with sleep modes and low-power sensors extending operational time between charges.
Looking ahead, the team sees potential for integrating artificial intelligence to enhance decision-making. For example, computer vision algorithms could allow the robot to identify plant species and adjust watering schedules accordingly. Machine learning models could predict water needs based on historical data, weather patterns, or seasonal changes. Future versions might also incorporate self-cleaning nozzles, automatic refilling from a central water source, or swarm coordination for large-scale deployments.
The publication of this research in Ordnance Industry Automation—a journal traditionally focused on defense and industrial systems—highlights the crossover potential of robotic technologies. Techniques developed for military or manufacturing applications are increasingly finding uses in civilian and consumer domains. This project exemplifies how engineering principles can be repurposed to address everyday challenges, bridging the gap between high-tech research and practical utility.
User feedback from early demonstrations has been overwhelmingly positive. Facility managers appreciate the labor savings, while plant enthusiasts value the consistency of care. One pilot test in a university atrium reported a 40% reduction in plant mortality over a three-month period after deploying the robot, compared to manual watering schedules. The system’s reliability and ease of use were cited as key factors in its success.
However, challenges remain. The current version requires a predefined path, limiting its ability to operate in completely unstructured environments. While the line-following method is effective, it is not as flexible as vision-based or LiDAR-guided navigation used in more advanced robots. Additionally, the robot’s size and weight may restrict access to certain areas, such as narrow aisles or elevated planters. Future iterations could explore smaller form factors or alternative mobility systems, such as tracks or legs, for greater versatility.
Another consideration is cost. While the researchers used commercially available components to keep expenses manageable, the total bill of materials—including sensors, microcontrollers, motors, and structural parts—still represents a significant investment for individual consumers. However, in commercial or institutional settings, the return on investment through reduced labor and improved plant health could justify the upfront cost.
Security and data privacy are also important, especially as the system relies on wireless communication. The team implemented basic encryption and authentication protocols in the NRF24L01 modules to prevent unauthorized access. Still, as with any connected device, ongoing vigilance is required to protect against potential cyber threats, particularly in networked building environments.
Despite these limitations, the mobile flower-watering robot represents a meaningful step forward in the evolution of domestic robotics. It combines proven technologies in a novel configuration, delivering tangible benefits in a niche yet widespread application. As smart homes and intelligent buildings become the norm, such specialized service robots will likely become standard fixtures, quietly maintaining the environments we live and work in.
The success of this project underscores a broader trend: the democratization of robotics. What was once the domain of industrial giants and research labs is now accessible to university teams and even hobbyists. Open-source hardware, affordable sensors, and powerful microcontrollers have lowered the barrier to entry, enabling innovation at the grassroots level. This robot, born from a collaboration between mechanical, electrical, and software engineers, exemplifies the interdisciplinary nature of modern robotics development.
In conclusion, the indoor mobile flower-watering robot developed by Lyu Pengfei and colleagues is more than just a gadget for plant lovers. It is a demonstration of how intelligent systems can enhance quality of life, optimize resource use, and reduce human burden in subtle but impactful ways. As the boundaries between robotics, environmental science, and human-centered design continue to blur, innovations like this will play an increasingly important role in shaping sustainable, livable spaces.
Indoor Mobile Flower-Watering Robot Developed by Lyu Pengfei et al. from Guangxi University of Science & Technology, published in Ordnance Industry Automation, DOI: 10.7690/bgzdh.2021.12.011