| Formatted contents note |
ABSTRACT: STATEMENT OF THE PROBLEM: There are gaps in the methods utilized for waste management, collection, and recycling as well as for its planning, diversion, and disposal preventing specific laws concerning waste management from being implemented correctly. Inconsistent and improper collection and segregation of waste is the main factor contributing to the waste being inappropriately segregated, which is the aim of the SegreBot system. As a whole, the researchers aim to develop a microprocessor-based plastic bottle and tin can segregation system utilizing image processing, and more specifically to: 1. To measure the accuracy of the object detection and classification system. 2. To integrate the system with a robotic arm that pick and sort the detected plastic bottles and tin cans into their respective bins. 3. To assess the usability of the system from the perspective of end-users, and to identify any potential issues related to usability or user experience. RESEARCH METHODOLOGY: This research study used a quantitative approach since the data came from the sensor’s detection status and the system’s functionality based on how well it can correctly segregate the waste materials. The data also came from a questionnaire regarding the system evaluation sent to the targeted participants. The overall project was tested at JBMond Junkshop, located at 1625 Paz St., Brgy 684, Zone 074 Paco, Manila. This junk shop produces a variety of solid waste, particularly plastic bottles and tin cans, which are essential for the execution of the system and from which the researchers may gather valuable information. The researchers used homogenous purposive sampling in selecting the participants, which is appropriate in quantitative research when the goal is to select participants with specific characteristics or experiences relevant to the research question. Aside from the questionnaire, the data also came from tests on the detection status of the sensors, and the robotic arm functions. SUMMARY OF FINDINGS: The researchers employed various data collection methods to achieve their specific objectives. The IR sensor proved to be effective in detecting waste on the conveyor belt, providing reliable results. The ultrasonic sensor successfully shut down the SegreBot system when the waste bin reached capacity and accurately indicated the waste level. The object detection model showed high average precision values for the “AluCan” (98.82%) and “PET” (87.94%) classes, demonstrating accurate detection. The model achieved a balance between precision and recall with an F1-score of 0.85, indicating overall effectiveness. The robotic arm exhibited a high level of precision is distinguishing plastic bottles and cans from mixed waste, with an average accuracy of 92.66%. However, there were instances where it missed picking up detected objects. The arm’s speed, averaging 10.31 cm/s, played a crucial role in its performance across the trials. The system usability scores were consistently high, with an average score of 93.21, indicating excellent usability and user satisfaction. The SegreBot system fulfilled the expectations and requirements of end users. CONCLUSION: Effective waste management is becoming an increasingly pressing issue for society, with plastic bottles and tin cans being two of the most generated types of waste. The SegreBot project aims to address this problem by developing a microprocessor-based plastic bottle and tin can segregation system that combines image processing and robotics. The study specifically had three (3) objectives in mind, alongside the said general objective. For the first specific objective, the group conducted a sensor detection test to know if the IR sensor and Ultrasonic sensor are working. They also tested the accuracy of the model through evaluation on a test data set. Since both tests gave positive results, the first specific objective had been met. The second objective deals with the function of the robotic arm. With several tests concerning the function of the robotic arm yielding feedback, the second objective had been met. With the last specific objective being concerned with participating feedback, the researchers conducted a survey questionnaire using the System Usability Scale as its basis for the questions. User feedback was overwhelmingly positive, with high levels of usability and satisfaction reported. In conclusion, the SegreBot project is a promising example of how technology can be used to solve real-world problems. RECOMMENDATIONS: The SegreBot project is a significant advancement in waste management technology. Recommendations for further development include replacing the robotic arm with a gantry system for improved efficiency and accuracy. Upgrading components like servo motors and implementing a customized gripper for specific waste types can enhance sorting precision. Using solar energy and powerful image processing tools would also increase sustainability and expand waste detection capabilities. Implementing the SegreBot system in schools, hospitals, and public areas would promote sustainable waste management, requiring additional research for adaptability. Overall, the SegreBot project has demonstrated the potential of technology to address pressing environmental challenges, particularly in waste management. By implementing the recommendations outlined above, the system’s efficiency, versatility, and sustainability can be further enhanced, making it a practical for waste management facilities worldwide.<br/> |