Hossein Rastgoftar

Matthew Romano, Prince Kuevor, Derek Lukacs et al.

This paper experimentally evaluates continuum deformation cooperative control for the first time. Theoretical results are expanded to place a bounding triangle on the leader-follower system such that the team is contained despite nontrivial tracking error. Flight tests were conducted with custom quadrotors running a modified version of ArduPilot on a BeagleBone Blue in M-Air, an outdoor netted flight facility. Motion capture and an onboard inertial measurement unit were used for state estimation. Position error was characterized in single vehicle tests using quintic spline trajectories and different reference velocities. Five-quadrotor leader trajectories were generated, and followers executed the continuum deformation control law in-flight. Flight tests successfully demonstrated continuum deformation; future work in characterizing error propagation from leaders to followers is discussed.

Hossein Rastgoftar, Ella M. Atkins

This paper proposes continuum deformation as a strategy for controlling the collective motion of a multiple quadcopter system (MQS) carrying a common payload. Continuum deformation allows expansion and contraction of inter-agent distances in a 2D motion plane to follow desired motions of three team leaders. The remaining quadcopter followers establish the desired continuum deformation only by knowing leaders positions at desired sample time waypoints without the need for inter-agent communication over the intermediate intervals. Each quadcopter applies a linear-quadratic-Gaussian (LQG) controller to track the desired trajectory given by the continuum deformation in the presence of disturbance and measurement noise. Results of simulated cooperative aerial payload transport in the presence of uncertainty illustrate the application of continuum deformation for coordinated transport through a narrow channel.

Hossein Rastgoftar, Ella Atkins

This paper studies collective motion optimization of a fleet of UAVs flying over a populated and geometrically constrained area. The paper treats UAVs as particles of a deformable body, thus, UAV coordination is defined by a homeomorphic continuum deformation function. Under continuum deformation, the distance between individual UAVs can significantly change while assuring the UAVs dont collide, enabling a swarm to travel through the potentially cluttered environment. To ensure inter-agent and obstacle collision avoidance, the paper formulates safety requirements as inequality constraints of the coordination optimization problem. The main objective of the paper is then to optimize continuum deformation of the UAV team satisfying all continuum deformation inequality constraints. Given initial and target configurations, the cost is defined as a weighted sum of the travel distance and distributed cost proportional to the likelihood of the human presence

Muhammad Junayed Hasan Zahed, Chieh Tsai, Salim Hariri et al.

This paper presents an online reinforcement-learning framework for safe gain scheduling of a nonlinear quadcopter controller. Rather than learning thrust and torque commands directly, the proposed method selects gain vectors online from a finite library of pre-certified stabilizing controllers, thereby preserving the structure of the underlying snap-based control law. Safety is enforced by restricting the policy to admissible gains that maintain forward invariance of a prescribed safe state set, while dwell-time constraints prevent excessively fast switching. To reduce the action-space dimension, translational gains are shared across spatial axes by exploiting the isotropic structure of the translational dynamics, whereas yaw gains are scheduled independently. A deep Q-network learns to adjust feedback authority according to the current flight condition, using aggressive gains during large transients and milder gains near hover. High-fidelity nonlinear simulations demonstrate accurate trajectory tracking, bounded attitude motion, reduced control effort near convergence, and stable hover regulation under online safe gain scheduling.

Chieh Tsai, Hossein Rastgoftar, Salim Hariri

Autonomous vehicles are increasingly deployed in safety-critical applications, where sensing failures or cyberphysical attacks can lead to unsafe operations resulting in human loss and/or severe physical damages. Reliable real-time perception is therefore critically important for their safe operations and acceptability. For example, vision-based distance estimation is vulnerable to environmental degradation and adversarial perturbations, and existing defenses are often reactive and too slow to promptly mitigate their impacts on safe operations. We present a Resilient Autonomous Car Framework (RACF) that incorporates an Object Distance Correction Algorithm (ODCA) to improve perception-layer robustness through redundancy and diversity across a depth camera, LiDAR, and physics-based kinematics. Within this framework, when obstacle distance estimation produced by depth camera is inconsistent, a cross-sensor gate activates the correction algorithm to fix the detected inconsistency. We have experiment with the proposed resilient car framework and evaluate its performance on a testbed implemented using the Quanser QCar 2 platform. The presented framework achieved up to 35% RMSE reduction under strong corruption and improves stop compliance and braking latency, while operating in real time. These results demonstrate a practical and lightweight approach to resilient perception for safety-critical autonomous driving

Chieh Tsai, Muhammad Junayed Hasan Zahed, Salim Hariri et al.

This paper proposes a safe reinforcement learning (RL) framework based on forward-invariance-induced action-space design. The control problem is cast as a Markov decision process, but instead of relying on runtime shielding or penalty-based constraints, safety is embedded directly into the action representation. Specifically, we construct a finite admissible action set in which each discrete action corresponds to a stabilizing feedback law that preserves forward invariance of a prescribed safe state set. Consequently, the RL agent optimizes policies over a safe-by-construction policy class. We validate the framework on a quadcopter hover-regulation problem under disturbance. Simulation results show that the learned policy improves closed-loop performance and switching efficiency, while all evaluated policies remain safety-preserving. The proposed formulation decouples safety assurance from performance optimization and provides a promising foundation for safe learning in nonlinear systems.

Hossein Rastgoftar

This paper presents a continuum mechanics-based approach for real-time deployment (RTD) of a multi-quadcopter system between moving initial and final configurations arbitrarily distributed in a 3-D motion space. The proposed RTD problem is decomposed into spatial planning, temporal planning and acquisition sub-problems. For the spatial planning, the RTD desired coordination is defined by integrating (i) rigid-body rotation, (ii) one-dimensional homogeneous deformation, and (ii) one-dimensional heterogeneous coordination such that necessary conditions for inter-agent collision avoidance between every two quadcopter UAVs are satisfied. By the RTD temporal planning, this paper suffices the inter-agent collision avoidance between every two individual quadcopters, and assures the boundedness of the rotor angular speeds for every individual quadcopter. For the RTD acquisition, each quadcopter modeled by a nonlinear dynamics applies a nonlinear control to stably and safely track the desired RTD trajectory such that the angular speeds of each quadcopter remain bounded and do not exceed a certain upper limit.

Harshvardhan Uppaluru, Hossein Rastgoftar

Mechanics of materials is a classic course of engineering presenting the fundamentals of strain and stress analysis to junior undergraduate students in several engineering majors. So far, material deformation and strain have been only analyzed using theoretical and numerical approaches, and they have been experimentally validated by expensive machines and tools. This paper presents a novel approach for strain and deformation analysis by using quadcopters. We propose to treat quadcopters as finite number of particles of a deformable body and apply the principles of continuum mechanics to illustrate the concept of axial and shear deformation by using quadcopter hardware in a $3$-D motion space. The outcome of this work can have significant impact on undergraduate education by filling the gap between in-class learning and hardware realization and experiments, where we introduce new roles for drones as "teachers" providing a great opportunity for practicing theoretical concepts of mechanics in a fruitful and understandable way.

Hossein Rastgoftar, Ella Atkins

This paper develops a novel physics-inspired traffic coordination approach and applies it to Unmanned Aircraft System (UAS) traffic management. We extend available physics-inspired approaches previously applied to 1-D traffic flow on highways and urban streets to support models of traffic coordination in higher dimension airspace for cases where no predefined paths exist. The paper considers airspace as a finite control volume while UAS coordination, treated as continuum deformation, is controlled at the airspace boundaries. By partitioning airspace into planned and unplanned spaces, the paper models nominal coordination in the planned airspace as the solution of a partial differential equation with spatiotemporal parameters. This paper also improves resilience to vehicle failures with a resilient boundary control algorithm to update the geometry of the planned space when UAS problems threaten safe coordination in existing navigable airspace channels. To support UAS coordination at the microscopic level, we propose clustering vehicles based on vehicle performance limits. UAS clusters, with each UAS treated as a particle of a virtual rigid body, use leader-follower containment to acquire the macroscopic desired trajectory.

Hossein Rastgoftar, Bingxin Zhang, Ella M. Atkins

This paper presents a novel data-driven approach to vehicle motion planning and control in off-road driving scenarios. For autonomous off-road driving, environmental conditions impact terrain traversability as a function of weather, surface composition, and slope. Geographical information system (GIS) and National Centers for Environmental Information datasets are processed to provide this information for interactive planning and control system elements. A top-level global route planner (GRP) defines optimal waypoints using dynamic programming (DP). A local path planner (LPP) computes a desired trajectory between waypoints such that infeasible control states and collisions with obstacles are avoided. The LPP also updates the GRP with real-time sensing and control data. A low-level feedback controller applies feedback linearization to asymptotically track the specified LPP trajectory. Autonomous driving simulation results are presented for traversal of terrains in Oregon and Indiana case studies.