Green Guardian - An IoT Solution for Plant Care

Recent Updates

May 19, 2024: Energy harvesting circuit added to prototype, test successful.

May 17, 2024: Code update for Wifi connection to Arduino, time synchronization, and entering low power mode.

May 13, 2024: Code update for reading sensor data of water level of actuator.

May 12, 2024: Code update for reading sensor data (Temperature, Humidity, Ambient Light, Soil Moisture).

May 8, 2024: Problem definition, product survey insights and technical approach added. Temperature and humidity sensor calibrated. Initial design of alogrithms.

April 24, 2024: Updated timeline and milestones. Created related work section.

April 20, 2024: Created product interest survey, please answer here!

Abstract

The Green Guardian project addresses substantial economic losses in the houseplant industry, where a significant portion of plants perish annually due to neglect. This IoT device simplifies plant care to a near-autonomous level, focusing primarily on precise soil humidity management through its integrated watering system. Once installed, Green Guardian monitors soil moisture and adjusts watering schedules accordingly, ensuring plants receive exactly what they need without risk of over or underwatering. It also monitors environmental factors such as light and temperature and advises owners when adjustments are needed to maintain optimal conditions. Powered by an energy-harvesting system, the device minimizes its environmental footprint and operates independently of external power sources. With its cloud-connected capabilities, Green Guardian offers users remote monitoring and minimal interaction, making it an ideal solution for ensuring plant health with little to no effort. This device not only promises to extend the life of houseplants but also to transform plant care into a hassle-free, sustainable routine.

Problem Definition

Americans invest approximately $16 billion annually in houseplants. Sadly, it is estimated that nearly half of these plants die due to improper care, which indicates a substantial gap in the consumer knowledge or available technology aiding in plant maintenance.

This situation presents a significant opportunity for innovation in the home gardening industry, particularly through the development of smart technologies that can reduce plant mortality rates and enhance user experience.

Milestones

Timeline

Week 1: Project kickoff, initial research, and component ordering are scheduled.

Week 2: Start of hardware assembly and basic software setup.

Week 4: Completion of the prototype assembly.

Week 5: Begin functional testing and iterate based on initial results.

Week 6: Development of the user interface and integration testing.

Week 7: Final testing, project documentation, and preparation for the demonstration.

Technical Approach: Details

The project proposes the creation of the Green Guardian, an IoT device designed to simplify plant care to a near-autonomous level. This device will manage watering schedules based on soil humidity data it collects, adjusting water delivery to the specific needs of each plant.

Implementation Diagram

Key components of the device include sensors for soil moisture, ambient light, and temperature, a water delivery system, and a power management unit featuring solar energy harvesting. This integrated system ensures that plants receive optimal care without requiring constant human intervention.

Product Survey - Preliminary Data

A survey was conducted with 119 participants to gauge potential user interest in the Green Guardian. Questions aimed to assess how often individuals interact with their plants and their interest in a self-sustaining plant care system.

The survey results for the Green Guardian product reveal promising insights into the market's readiness and potential interest in an autonomous plant care solution. A majority of respondents interact with their plants frequently (daily or several times a week), indicating an existing commitment to plant care that the Green Guardian could support and enhance. Additionally, a significant number of respondents reported having plants that died prematurely, suggesting a prevalent challenge in maintaining plant health that the Green Guardian aims to address. Importantly, the strong inclination towards a self-contained watering system, with a substantial majority expressing interest, underscores the demand for such innovative solutions. This interest, coupled with respondents' willingness to invest in the price range of $20 to $60, positions the Green Guardian as a viable product in the houseplant care market.

Survey response chart1 Survey response chart2 Survey response chart3 Survey response chart4 Survey response chart5 Survey response chart6

Sensor Selection, Fusion and Calibration

Environmental factors that affect plant growth and include light, temperature, water, humidity and nutrition.​[16] The first four parameters: light, temperature, humidity and soil moisture (through watering) are readily measurable using cost-effective off-the-shelf sensors, which we procured. Nutrition is tougher to sense and requires more expensive approaches such as spectroscopy, wifi signal, pH, ion-selective electrodes.​ Since our design goals require an affordable solution, we decided to forgo the pH sensor (cheapest option would cost $70).

As we do not want affordability of the system to compromise the quality, we needed to calibrate the sensors so that the data they produce is reliable. We began with calibrating our DHT11 temperature and humidity sensor. The below method involved taking initial readings from the sensor and comparing it to the temperature and humidity values as measured by an iPhone. We then calibrated the sensor readings to the correct value through one-stop calibration. The sensor calibration code is available on our github repository.

Plant Care through Actuation

Our goal is to take care of a plant with minimal or no human effort. To this end, we decided to design an actuation system that would water the plant as and when directed by a decision making model. This decision making model is designed by utilizing the plant specific care instructions, along with the sensor data for soil moisutre and ambient light.

We define some terms for our system:

  • Environment: Plant + data collected by sensors​
  • State: Current conditions of the plant defined by data collected from soil moisture, temperature, humidity, and ambient light sensors.​
  • Agent: Arduino Controller
  • Action: Water the plant / Do not water the plant​
  • Initial Approach: Rule Based Decision Making

  • Use thresholds based on plant specific care instructions: watering frequency, light requirements, other conditions (e.g., water after soil is completely dried out).
  • Do not water plants at night.​ Because of absence of sunlight, leaves do not dry off and are prone to fungal infestations.
  • Water specific amount corresponding to plant care instructions.​
  • Advanced Approach: Actuation-with-Learning

  • Develop a comprehensive dataset from plant sensor data collected over several days.​
  • Establish a correlation between the plant's condition and actuation actions, focusing on watering frequency and volume.​
  • Utilize this "learning" approach to create a plant-specific model.​
  • Enhance decision-making processes by optimizing factors such as thresholds and watering parameters.​
  • Rule Based Actuation
    Actuation second diagram
    Actuation with Learning

    Energy Autonomy through Energy Harvesting

    Our system goals are self-sufficiency, long lifespan, and energy efficiency. However, the system components are battery powered and will get depleted over time.​ Therefore, we design an energy harvesting system to power GreenGuardian as a stand-alone, long-term solution.
  • Perform energy profiling to realize energy consumption.​
  • Harvest Solar Energy to power system.​
  • Apply optimization techniques.​
  • Energy harvesting component specifications

    Optimization Techniques

  • Have separate power sources for sensing module and control module.​ - Gives software control over sensor activation and operation.
  • Sensor data is not highly variable, so we are not sensing continuously​. Sensors are turned off and Arduino is in low-power mode for majority of the time.​
  • Adaptive Duty Cycling ​- Battery is charged on good weather days, and sensor activity is lessened on bad weather days (Low Battery Level + Ambient Light below threshold)​. This enables us to adapt duty cycling to battery power and availability of ambient light.​