Benchmarking Tracking Autopilots for Quadrotor Aerial Robotic System Using Heuristic Nonlinear Controllers

Abstract

This paper investigates and benchmarks quadrotor navigation and hold autopilots’ global control performance using heuristic optimization algorithms. The compared methods offer advantages in terms of computational effectiveness and efficiency to tune the optimum controller gains for highly nonlinear systems. A nonlinear dynamical model of the quadrotor using the Newton–Euler equations is modeled and validated. Using a modified particle swarm optimization (MPSO) and genetic algorithm (GA) from the heuristic paradigm, an offline optimization problem is formulated and solved for three different controllers: a proportional–derivative (PD) controller, a nonlinear sliding-mode controller (SMC), and a nonlinear backstepping controller (BSC). It is evident through the simulation case studies that the utilization of heuristic optimization techniques for nonlinear controllers considerably enhances the quadrotor system response. The performance of the conventional PD controller, SMC, and BSC is compared with heuristic approaches in terms of stability and influence of internal and external disturbance, and system response using the MATLAB/SIMULINK environment. The simulation results confirm the reliability of the proposed tuned GA and MPSO controllers. The PD controller gives the best performance when the quadrotor system operates at the equilibrium point, while SMC and BSC approaches give the best performance when the system does an aggressive maneuver outside the hovering condition. The overall final results show that the GA-tuned controllers can serve as a benchmark for comparing the global performance of aerial robotic control loops.

Surrogate Optimal Fractional Control for Constrained Operational Service of UAV Systems

“…Heuristics and meta-heuristics techniques have also been proposed in the literature. In particular, the genetic algorithm, the particle swarm optimization, the duelist algorithm, the simulated annealing, the ant colony optimization, and the Manta-ray foraging optimizer are examples of meta-heuristic algorithms [5]. Furthermore, the acquisition of analytical solutions for problems involving fractional orders is not universally feasible, due to the intricate nature of the associated mathematical expressions.…”

“…A guidance abstract model for a 1 kg quadrotor UAV was utilized to investigate the tuning of the FOPID using surrogate optimization. The model was a scaled version of the quadrotor described in detail in [5]. The kinematic model was acquired from the model developed in [66] and was implemented in the MATLAB UAV toolbox [67], incorporating a linear dynamical model controlled by orientation and thrust hold autopilot.…”

“…The kinematic model was acquired from the model developed in [66] and was implemented in the MATLAB UAV toolbox [67], incorporating a linear dynamical model controlled by orientation and thrust hold autopilot. The notations and conventions for quadrotor modeling, available in [5,66], are given in this subsection. The quadrotor used two frames: the earth frame [x, y, z] and the body frame [x b , y b , z b ].…”

“…In contrast to conventional UAVs, aerial robot manipulators such as quadrotors and twin rotors present distinct kinematic architectures. For instance, a quadrotor employs a unique configuration with two rods and four actuators, achieving torque equilibrium by coordinating two rotors to rotate clockwise and the remaining two to rotate counterclockwise [5]. Similarly, a twin-rotor design operates on a comparable principle.…”

Rapid trajectory generation for quadrotor with limited motor speed

“…[21][22][23] For autonomous UAV flight, the tracking controller is in the bottom layer to ensure UAV flight safety, and fault-tolerant control have been widely utilized for fault tolerance. 24 The mainstream idea to realize compensation of speed and accurate tracking of reference instruction is based on controller reconfiguration, but this method is only applicable to the case of slight decrease of motor speed. This treatment is not effective when the upper speed limit of the motor is greatly reduced and cannot provide sufficient execution capacity.…”