This paper presents the design, implementation, and experimental validation of a compact Ackermann-steered mobile robot platform developed for autonomous navigation and diverse service applications. The platform integrates dual stereo cameras and an inertial measurement unit to ensure robust localization, alongside a pan-tilt RGB-D camera designed for active perception tasks. Specifically, the dual stereo cameras mounted at the front and rear provide bidirectional visual odometry, which is fused with inertial measurements via an extended Kalman filter to achieve accurate, drift-reduced state estimation. Furthermore, the pan-tilt actuator enables dynamic viewpoint control, facilitating robust object tracking and dense three-dimensional (3D) perception. The system is built upon the Robot Operating System 2 framework, providing a modular and extensible architecture that allows for seamless integration with various perception and navigation algorithms. Experimental evaluations—including velocity tracking, dual visual odometry benchmarking, mapless navigation, and real-time 3D mapping—validate the effectiveness and versatility of the proposed platform in outdoor environments.
Han et al. (2026) studied this question.