Robot Arm Technology in Detection and Manipulation in Smart Pharmacies: A Review

Abstract

Detection and manipulation are critical functions in robotic arm technology within smart pharmacy environments, enabling accurate and efficient handling of pharmaceutical products. Despite rapid advancements, challenges persist in visual recognition under varying conditions, hardware–software integration, and real-time responsiveness. This systematic review, conducted using the PRISMA methodology, analyzes 20 selected articles from reputable databases such as MDPI, Wiley, ScienceDirect, IEEE Xplore, and others. The findings highlight the integration of advanced sensors such as force, torque, tactile, and RGB-D, and the application of deep learning models such as YOLO, CNN, and Mask R-CNN, which collectively improve object detection and manipulation accuracy. Reinforcement learning and classical AI planning further support adaptive automation. Future research directions include hybrid AI fuzzy logic models, edge computing for real-time data processing, digital twin simulations for predictive optimization, and micro–nano robotic technologies for precision drug handling. These innovations are expected to enhance the safety, reliability, and performance of robotic systems in pharmaceutical automation.