Design of an Intelligent Following Stroller System Based on Computer Vision

Aiming at the drawbacks of single functionality in existing strollers and the lack of convenience in infant care and mobile travel, this study focuses on the integrated innovation of computer vision and embedded technology to develop an intelligent stroller system. Through the innovative integration of infant posture detection and stranger detection modules, the system significantly enhances environmental perception capabilities and the accuracy of infant status monitoring, thereby enabling real-time activation of safety protection mechanisms. The hardware architecture uses STM32 as the main control core and relies on the MQTT protocol to build a cloud-based IoT platform, which achieves efficient data interaction between terminals and the cloud and constructs an intelligent interactive interface on mobile terminals that integrates data monitoring and device control functions. Field test results demonstrate that the system operates stably and reliably, effectively reducing the parental effort required for infant care, meeting the needs of multiscenario infant care and travel, and providing an innovative solution for intelligent parenting.