What it does
The system monitors wind, soil moisture, and tree size to assess stability. Using an Arduino controller and AI, it detects high-risk trees and gives timely alerts to reduce the risk of tree falls in urban areas.
Your inspiration
The idea was sparked by a storm on 25 May 2024 at Universiti Malaya, where multiple trees fell, damaging vehicles and blocking access. This highlighted the urgent need for a system to assess and mitigate tree fall risks. Reports from the Malaysian Civil Defence Force showed rising incidents of tree falls in recent years, often due to aging trees, poor maintenance, and unpredictable weather. Our team aimed to create a data-driven, sustainable solution that uses real-time monitoring and AI to prevent such incidents and improve urban safety.
How it works
The system works by collecting real-time environmental data and analysing tree-specific inputs to assess fall risk. An anemometer measures wind speed, while soil moisture sensors monitor ground conditions. Tree measurements such as height, canopy area, and trunk diameter are taken manually, and the canopy area is calculated using our custom-developed software from a reference image. These values are entered through a 4x4 keypad into an Arduino Mega, which acts as the system’s processor. It calculates drag force on the tree and compares it against soil stability to estimate the likelihood of failure. If the tree is at high risk of falling, the system activates a buzzer, lights a red LED, and displays a warning on a 2.4-inch OLED screen. A green LED indicates safe conditions. The entire system is powered by a solar panel and a power management module, allowing continuous outdoor operation without external electricity.
Design process
We began by researching the growing issue of tree falls in Malaysia and conducted a market survey involving 113 participants to understand user needs. Based on the data, we generated six concept designs and used a decision matrix to evaluate them. The highest-scoring design was selected: a fixed system using sensors and a centralised unit. We developed detailed CAD models for the tower housing and internal layout. In Semester 2, we purchased components such as the Arduino Mega, soil moisture sensors, anemometer, OLED display, solar panel, and power modules. We assembled and wired the system on a breadboard, then developed custom software to calculate canopy area from tree images. Testing began with individual sensors, followed by integration and calibration to ensure reliable readings. We iterated on the code to refine risk calculations based on drag force and soil data. Visual outputs and warning systems were added, including LEDs, buzzer, and display. The final prototype is a fully functional, solar-powered unit capable of assessing tree fall risk in real time.
How it is different
Our system stands out by combining real-time environmental monitoring, AI prediction, and custom software in a single, fixed-location unit. Unlike conventional methods like manual inspections, which are time-consuming and subjective, or advanced diagnostic tools like sonic tomography, which are costly and require expert handling, our design provides continuous, automated assessment. It uses wind and soil sensors along with manually measured tree parameters to calculate drag force and root stability. We developed our own software to calculate canopy area from tree images, removing the need for third-party analysis tools. The system is solar-powered, low-maintenance, and gives instant feedback through visual displays and alerts. Designed for long-term deployment in campuses and urban zones, it offers a practical and scalable solution for improving public safety in response to unpredictable weather events.
Future plans
We plan to expand the system with a cloud-based dashboard for remote monitoring, risk trend analysis, and automated report generation. Features like GPS tagging and real-time heatmaps will visualise high-risk areas. The AI model will be trained with more data to improve prediction accuracy. We also aim to add wireless alerts for maintenance teams when trees reach critical risk levels. Future deployments in urban zones, campuses, and parks will be tested in collaboration with local councils to support proactive tree management and public safety.
Share this page on