GestureBot AI : Smart Surveillance Car

This research successfully designed, implemented, and val-idated the GestureBot AI, an advanced smart surveillance car engineered explicitly to modernize hazardous waste inspec-tion. By synthesizing intuitive gesture kinematics with a highly optimized, single-stage deep learning classification pipeline, the system actively and physically decouples vulnerable hu-man sanitation workers from acute occupational hazards. The implementation of a strictly isolated dual-channel telemetry architecture guarantees phenomenal mechanical re-sponsiveness without succumbing to inevitable video band-width interference. Achieving an overall classification accu-racy of 95.7% with a critical 94.3% sensitivity to hazardous materials, the prototype confirms the absolute viability of integrating cost effective microcontrollers with state of the art artificial intelligence. This research establishes a scalable, deeply effective framework for human-machine collaboration, offering a highly practical technological solution to modern municipal solid waste management challenges. Future itera-tions will focus on transitioning the visual processing entirely to the edge device and integrating multi axis manipulators to evolve the platform from a passive surveillance tool into an active, autonomous segregation unit.

[1]     S. Sakib, M. Toha, J. A. Adreeta, and M. F. Muhtashim, ”Hand Gesture Controlled Robotic Car Using ESP32,” American International University-Bangladesh, Dhaka, Bangladesh, 2024. [2]     T. Diwan, G. Anirudh, and J. V. Tembhurne, ”Object detection using YOLO: challenges, architectural successors, datasets and applications,” Multimedia Tools and Applications, vol. 82, pp. 9243-9275, 2023. [3]     C. K. Wen and L. C. Yong, ”Gesture Controlled Robotic Arm For Hazardous Chemical Control,” JATI Journal of Applied Technology and Innovation, vol. 6, no. 3, pp. 47-55, 2022. [4]     T. Dorigo et al., ”Artificial Intelligence in Science and Society: The Vision of USERN,” IEEE Access, vol. 13, 2025. [5]     I. Ahammed, M. K. A. Rakib, M. Y. Rahman, J. Mawa, and A. Obaidah, ”Edge-AI Vision Surveillance Robot: Real-Time Object Detection and IoT-Driven Autonomous Navigation,” in Proc. 7th Global Power, Energy and Communication Conference (GPECOM), Bochum, Germany, 2025. [6]     S. A. Bowyer and F. Rodriguez y Baena, ”Dissipative Control for Physical Human-Robot Interaction,” IEEE Transactions on Robotics, vol. 31, no. 6, pp. 1281-1293, Dec. 2015. [7]     C. Herkariawan, N. R. S. Muda, D. Minggu, E. Kuncoro, R. Agustiady, and M. L. N. Suryana, ”Design control system using gesture control on the Arduino-based robot warfare,” IOP Conference Series: Materials Science and Engineering, vol. 1098, 032013, 2021. [8]     R. Setiawan, H. H. Triharminto, and M. Fahrurozi, ”Gesture Control Menggunakan IMU MPU 6050 Metode Kalman Filter Sebagai Kendali Quadcopter,” Prosiding Seminar Nasional Sains Teknologi dan Inovasi Indonesia, vol. 3, pp. 411-422, 2021. [9]     L. M. Xin and N. F. N. Rosely, ”Development of Machine Learning Wildlife Camera Using ESP32 CAM for Small Mammals,” Pertanika J. Sci. & Technol., vol. 33, no. 4, pp. 1783-1807, 2025. [10]    M. Meiswinkel, A. Ridder, and A. Ebrahimi, ”Contactless Telemetry System for Condition Monitoring of Large Hydro and Wind Generators,” IEEE Access, 2026. [11]    J. Sun and L. Qiu, ”Human-Machine Collaboration: Definitions, Models, and Socio-Economic Impacts,” International Journal of Crowd Science, 2025. [12]    P. A. Rajeev et al., ”Advancing e-waste classification with customizable YOLO based deep learning models,” Scientific Reports, vol. 15, no. 18151, 2025.