Abstract

Sliding Mode Control (SMC) is a robust and adaptable control strategy recognized for its ability to withstand uncertainties, disturbances, and parameter variations. Although the straightforward implementation of SMC along with its insensitivity to modeling inaccuracies make it effective for a wide range of applications, it still relies heavily on an accurate system model. In dynamic environments where system characteristics vary over time, a model-independent SMC framework becomes necessary. So, a Model-Free Sliding Mode Control (MFSMC) approach is proposed as a compelling solution for complex nonlinear systems where deriving an accurate model is difficult. In contrast to conventional SMC approach, MFSMC requires only the system’s state measurements, order, and previous control inputs, enabling it to adapt to highly complex and unpredictable operating conditions. Most practical systems are inherently underactuated. To address this issue, MFSMC is combined with a hyperplane transformation, which reduces the system states so that they correspond to the available control inputs. The hyperplane transformation matrix is also estimated in real time to enhance overall system performance. Finally, this work also addresses the implementation of MFSMC for quadrotor systems. Indoor UAVs are increasingly employed in search and rescue missions, where precise mapping and reliable waypoint following, are essential for operating in confined, hazardous, and GPS-denied environments. However, lightweight platforms such as the DJI Tello often suffer from substantial drift because of limited onboard sensors and low-grade inertial data, which degrades SLAM performance. So, a visual SLAM framework with a discrete model-free sliding mode controller (MFSMC) is introduced in this work to achieve robust trajectory tracking and counteract drift. The proposed method allows the drone to accurately follow predefined paths while producing stable and accurate environmental maps. Experiments conducted on the Tello platform show enhanced tracking precision, reduced drift buildup, and reliable waypoint following in indoor settings. These findings underscore the promise of SMC-augmented UAVs for real-time mapping and navigation in time-sensitive rescue operations.

Publication Date

2-18-2026

Document Type

Dissertation

Student Type

Graduate

Degree Name

Engineering (Ph.D.)

Department, Program, or Center

Engineering

College

Kate Gleason College of Engineering

Advisor

Agamemnon Crassidis

Advisor/Committee Member

Jason Kolodziej

Advisor/Committee Member

Kathleen Lamkin-Kennard

Campus

RIT – Main Campus

Download

Share

COinS