This paper introduces a novel control strategy for robotic obstacle avoidance by integrating Gradient Tracking Integral Sliding Mode Control (GT-ISMC) with Artificial Potential Fields (APFs). This approach aims to enhance autonomous navigation's robustness and accuracy in dynamic and uncertain environments. Traditional obstacle avoidance methods struggle with local minima and lack sufficient robustness against external disturbances and model uncertainties. The proposed method employs APFs to create smooth repulsive forces to avoid obstacles and attract the robot to the target. The novelty lies in designing an integral sliding mode controller to track the APF gradients, improving robustness against disturbances and uncertainties in the robot's dynamics, reducing chattering and improving convergence performance. Stability analysis guarantees the convergence of the robot to the desired reference while evading the obstacle. Simulations and experimental results demonstrate that the GT-ISMC with APFs effectively navigates complex environments, offering superior robustness, faster convergence, and smoother trajectories compared to conventional PID, sliding mode, and Potential Field methods. This work contributes to the advancement of autonomous robotic control by providing a reliable and efficient method for real-time obstacle avoidance, with potential applications in mobile robotics and industrial automation. Highlights A novel control strategy for robotic obstacle avoidance by integrating Gradient Tracking Integral Sliding Mode Control (GT-ISMC) with Artificial Potential Fields (APFs) is proposed. The controller enhances the robustness and accuracy of autonomous navigation in dynamic and uncertain environments, where some motion constraints are considered. The proposed method employs APFs to generate smooth, continuous repulsive forces that guide the robot away from obstacles while attracting it toward the target. By incorporating gradient tracking into the sliding mode framework, the system benefits from reduced chattering and improved convergence performance.
