This work focuses on the design and development of an intelligent pick and place robotic arm system using computer vision and machine learning with embedded control. A Raspberry Pi 3B+ is used as the main controller, which carries out real-time image processing with OpenCV and a lightweight CNN-based object detection model. A Raspberry Pi Camera is utilized to offer a real-time video stream enabling the recognition of the objects by their visual traits and executing the automatic pick and place process. The robotic arm driven by 5-DOF servo and controlled by PCA9685 servo driver, the Raspberry Pi coordinate to fulfill the accurately grabbing and placing. The system is mounted on a mobile rover base driven by DC gear motors, with flask-based Wi-Fi interface for remote monitoring and manual intervention through any smart gadgets. This up-conversion from 2 column to 1 column makes it easier to read the correct system identification, which clearly demonstrates that the proposed system provides an effective and inexpensive solution for AI enabled automation and could be utilized in material handling, sorting, and academic research as well as laboratory training. Its modularity serves as a basis for future enhancements such as advanced navigation, multi-object detection, and IoT-based remote operations.

This work provides a convincing proof-of-concept for a smart pick and place robotic system that combines computer vision, real-time embedded control, and servo-based actuation to realize an autonomous platform. Based on CNN object detection and real-time image processing, the robot locates objects, recovers visual information, and executes synchronized pick and place operations with little user interaction. It illustrates how low-cost hardware and open-source software is sufficient to deliver capabilities traditionally available only with high end industrial robots. The work done is highly successful and meets all the goals set out for the project, the outcome being a practical, modular and scalable platform upon which to build accessible AI driven automation. The design also allows for future enhancements in navigation, perception and IoT which ultimately can lead to the development of a mobile robot suitable for academic and small industrial uses. In summary, the project shows that intelligent robotics can be made cheap, flexible, and educationally rewarding.

[1] N. S. K. Rao, Avinash, R. Moorthy H., Karthik K., S. Rao, and S. Santosh, “An automated robotic arm: A machine learning approach,” in Proc. IEEE Int. Conf. Mobile Networks and Wireless Communications (ICMNWC), 2021. [2] M. Abdelaal, “A study of robot control programming for an industrial robotic arm,” Computers and Systems Department, Electronics Research Institute, Giza, Egypt. [3] R. Yenorkar and U. M. Chaskar, “GUI based pick and place robotic arm for multipurpose industrial applications,” in Proc. 2nd Int. Conf. Intelligent Computing and Control Systems (ICICCS), 2018, pp. —. IEEE Xplore compliant, ISBN: 978-1-5386-2842-3. [4] A. Kumar and S. Majumder, “Vision-based object detection and recognition for robotic manipulation,” IEEE International Conference on Robotics and Automation (ICRA), 2019, pp. 455–462. [5] R. J. Hadsell, P. Nguyen, and L. Pinto, “Deep learning-based grasp detection using CNNs for real-time robotic manipulation,” IEEE Robotics and Automation Letters, vol. 4, no. 2, pp. 234–241, 2019. [6] J. M. Romano, K. Hsiao, and K. Kuchenbecker, “Human-inspired robotic grasp control using tactile and force feedback,” IEEE Transactions on Robotics, vol. 27, no. 6, pp. 1067–1079, 2011. [7] S. Thrun, M. Montemerlo, and W. Whittaker, “Vision-based mobile robot localization and mapping,” IEEE Robotics & Automation Magazine, vol. 13, no. 2, pp. 80–91, 2016. [8] S. Zheng, J. Wang, and X. Zhao, “CNN-based object classification and detection using Raspberry Pi for real-time applications,” in Proc. IEEE Int. Conf. Intelligent Computing and Control Systems (ICICCS), 2020, pp. 110–115. [9] T. Pereira and M. Veloso, “A mobile platform with vision-based navigation and object interaction,” in Proc. IEEE/RSJ Int. Conf. Intelligent Robots and Systems (IROS), 2018, pp. 211–218. [10] S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards real-time object detection with region proposal networks,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137–1149, 2017. [11] D. Kijdech and S. Vongbunyong, “Manipulation of a complex object using dual-arm robot with Mask R-CNN and grasping strategy,” Journal of Intelligent & Robotic Systems, vol. 110, article 103, Jul. 2024. [12] E. Govi et al., “A deep-learning-based six degrees-of-freedom pose estimation for robotic pick-and-place,” in Proc. Robotics Conference, 2024. [13] H. A. Khan et al., “Design and development of machine-vision robotic arm for automated fruit picking,” ScienceDirect, 2024. [14] J. Galarza-Falfán et al., “Path planning for autonomous mobile robot using convolutional neural networks,” Technologies, vol. 12, no. 6, 2024. [15] A. Efendi, “Robot arm technology in detection and manipulation: Review of recent sensors, machine learning and applications,” World Scientific Review, 2025. [16] R. Dasarwar, N. Thakre, K. Tikale, A. Dhoke, and T. Randive, “Industrial robotic arm using Raspberry Pi,” Int. J. Innovations in Engineering and Science, vol. 9, no. 3, pp. 44–50, May 2024. doi: 10.46335/IJIES.2024.9.3.9. [17] Raspberry Pi Foundation, “Raspberry Pi Documentation,” Raspberry Pi Ltd. [Online]. Available: https://www.raspberrypi.com/documentation/. Accessed: 2025. [18] Raspberry Pi Foundation, “Connect to a Raspberry Pi from another computer,” Raspberry Pi Ltd. [Online]. Available: https://www.raspberrypi.com/software/connect/. Accessed: 2025. [19] Arduino, “Arduino Uno Reference,” Arduino S.r.l. [Online]. Available: https://www.arduino.cc/reference/en/ Accessed: 2025. [20] TensorFlow, “TensorFlow Documentation,” Google Brain Team. [Online]. Available: https://www.tensorflow.org/. Accessed: 2025.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close