Real-Time Leader-Follower UGV-UAV Cooperation: Experimental Validation and Control Analysis

Abstract

Abstract This paper presents the experimental validation of a real-time leader–follower cooperative control architecture for a heterogeneous system composed of a differential-drive Unmanned Ground Vehicle (UGV) and a quadrotor Unmanned Aerial Vehicle (UAV). In the proposed framework, the ground robot acts as the leader and generates the system motion through trajectory tracking, while the aerial robot follows the leader based exclusively on its instantaneous state transmitted in real time, without access to any predefined reference trajectory. This establishes a state-based leader–follower strategy that differs from classical trajectory-synchronization approaches. Two nonlinear control strategies, Sliding Mode Control (SMC) and Integral Backstepping, are designed and experimentally compared for the leader robot to ensure robust and accurate trajectory tracking under real-world conditions. The follower UAV employs a Proportional–Integral–Velocity (PIV) controller, which reconstructs its reference motion online from the received leader state. An OptiTrack motion capture system is used solely for state estimation and does not provide trajectory references to the follower, ensuring that the cooperative behavior arises from real-time state exchange. The complete control architecture is implemented in MATLAB/Simulink and executed in real-time on a ground station, with control commands wiressly transmitted to the QBot 2e and QDrone platforms. Experimental results obtained on circular and lemniscate trajectories demonstrate accurate tracking, reliable real-time coordination, and consistent leader–follower behavior. The comparative analysis shows that Integral Backstepping provides smoother control actions and reduced oscillations compared to SMC, resulting in stable reference generation for the follower. These results highlight the practical feasibility of state-based cooperative control in heterogeneous robotic systems and provide an experimentally validated framework for real-time UGV–UAV leader–follower interaction.