Modern aircraft have several safety features to ensure that all flight parameters remain in their normal operating range. However, if the aircraft diverges from desired conditions, it enters an upset condition in whic...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
Modern aircraft have several safety features to ensure that all flight parameters remain in their normal operating range. However, if the aircraft diverges from desired conditions, it enters an upset condition in which the flight dynamics are highly nonlinear and integrity of the aircraft is endangered. In upset conditions, nonlinear control laws are hence necessary to handle the nonlinear dynamics. Verification of such nonlinear flight control laws is not a trivial task as common verification methods only apply to linear control systems. The contribution of this paper is to formulate a scenario optimisation method for verification of an exemplary, switching-based upset recovery control law. In addition, we propose a two-step optimisation approach to solve the problem efficiently. Our method is applicable to arbitrary nonlinear control laws given that a simulation model is available. Here, simulations were conducted using the Generic Transport Model, a high-fidelity flight dynamics simulation developed by NASA.
We propose algorithms for controlling an articulated snake-like robot using a 2-axis joystick. The key contributions of the paper are two-fold: i. Development of a pipeline for converting the joystick inputs into Cart...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
We propose algorithms for controlling an articulated snake-like robot using a 2-axis joystick. The key contributions of the paper are two-fold: i. Development of a pipeline for converting the joystick inputs into Cartesian velocity commands for the robot's end-effector in a way that is intuitive to a user, and, ii. Development of an optimization-based controller that generates the necessary joint velocities to track the target Cartesian velocities as closely as possible even at or near singular configurations of the robot. We describe the proposed framework and outline its technical/mathematical details. We also demonstrate the effectiveness of the proposed approach through Gazebo simulations as well as real-robot experiments using a prototype of a snake-like robot called FLX BOT.
This study examines a robust control approach for a floating offshore wind turbine (FOWT) operating in Region II. The aim is to maximize power point tracking (MPPT) while minimizing fatigue loads within wind speed lim...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
This study examines a robust control approach for a floating offshore wind turbine (FOWT) operating in Region II. The aim is to maximize power point tracking (MPPT) while minimizing fatigue loads within wind speed limits. Ensuring the tip speed ratio (TSR) is at its optimal value achieves the MPPT control aim while setting the platform pitch velocity to zero helps in reducing system fatigue. control design based on the OpenFAST platform model is not trivial due to the nonlinear and complex dynamics provided by the model of a FOWT. A robust nonlinear control approach is proposed to address this issue as it requires minimal knowledge about the model of the system. Such a class of controllers is adapted to parameter variations and perturbation. Finally, simulation results are presented to validate the effectiveness of the implemented control approaches.
This paper presents a safe control design for systems with unknown nonlinear dynamics, using a proposed adaptive recurrent neural network. For safety-critical control, several noteworthy results and methods have been ...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
This paper presents a safe control design for systems with unknown nonlinear dynamics, using a proposed adaptive recurrent neural network. For safety-critical control, several noteworthy results and methods have been presented that not only stabilize affine control systems to desired states but also ensure satisfaction of state and control constraints. One of the main challenges for safety-critical control design is the accurate knowledge of system dynamics, which is challenging to obtain for many practical systems subjected to limited computational resources and historical data. To address the fore-mentioned problem, we first present an adaptive recurrent neural network (AdaRNN) which leverages the concept of continual learning to learn and adapt to unknown nonlinear system dynamics with limited resources, followed by the utilization of AdaRNN outputs for the design of a control barrier function based quadratic program, to ensure safe behavior of the system. Finally, we illustrate the effectiveness of the presented approach through both field and simulation data.
In this paper we consider data-driven control of vibrations in a flexible structure equipped with piezoelectric transducers. The control algorithm uses elements from the fairly recent data-enabled predictive control f...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
In this paper we consider data-driven control of vibrations in a flexible structure equipped with piezoelectric transducers. The control algorithm uses elements from the fairly recent data-enabled predictive control framework. In particular, we will develop a real-time implementation of the subspace predictive control algorithm. This algorithm first solves a linear least-squares problem to recursively estimate the observer Markov parameters of the system. With those parameters a predictor is constructed which is used to solve a predictive control problem subject to constraints. The feasibility of the approach is highlighted by applying it to an experimental setup using an efficient implementation. First, this demonstrates that computations can be performed in real-time for a realistic situation. Second, we show how the scheme rapidly adapts when a sudden significant change in structural dynamics is introduced by changing one of the structural parameters.
The control Lyapunov function (CLF) approach to nonlinear control design is well established. Moreover, when the plant is control affine and polynomial, sum-of-squares (SOS) optimization can be used to find a polynomi...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
The control Lyapunov function (CLF) approach to nonlinear control design is well established. Moreover, when the plant is control affine and polynomial, sum-of-squares (SOS) optimization can be used to find a polynomial controller as a solution to a semidefinite program. This letter considers the use of data-driven methods to design a polynomial controller by leveraging Koopman operator theory, CLFs, and SOS optimization. First, Extended Dynamic Mode Decomposition (EDMD) is used to approximate the Lie derivative of a given CLF candidate with polynomial lifting functions. Then, the polynomial Koopman model of the Lie derivative is used to synthesize a polynomial controller via SOS optimization. The result is a data-driven method that skips the intermediary process of system identification and can be applied widely to control problems. The proposed approach is used to successfully synthesize a controller to stabilize an inverted pendulum on a cart.
This paper proposes a field-of-view (FoV) assistive control technique called redirected walking (RDW) that smoothly links human motion in the real world and virtual reality (VR) space by maneuvering through motion ill...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
This paper proposes a field-of-view (FoV) assistive control technique called redirected walking (RDW) that smoothly links human motion in the real world and virtual reality (VR) space by maneuvering through motion illusion. In particular, to improve the cooperative consensus control between the human operator equipped with the head mount display in the real world and the drone swarm in VR space, one technical issue is to apply the RDW technique to a dynamical formulation implementable in real-time. To overcome the issue, we propose a dynamic RDW technique based on sigmoid gain mapping. To evaluate the quantitative performance of the proposed control method, we have created a new experimental environment with Unreal Engine 4, AirSim, Meta Quest3, and Python. The effectiveness of the proposed method is finally demonstrated through the developed experimental environment. As a result, we see that the proposed method can improve the drone swarm's tracking and maneuvering performance and suppress discrepancies between somatosensory perception and the human operator's FoV.
This paper considers the design of an attitude controller of a launcher upper stage module during its exo-atmospheric phase, where short boosts are performed to adapt the flight path. Those maneuvers can increase prop...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
This paper considers the design of an attitude controller of a launcher upper stage module during its exo-atmospheric phase, where short boosts are performed to adapt the flight path. Those maneuvers can increase propellants' motion in the tanks, leading to the so-called sloshing phenomenon that may affect the stability of the vehicle. A Deep Reinforcement Learning (DRL) algorithm is proposed to design the controller accounting for non linearities of the launcher and sloshing dynamics as well as presence of time-delay, bias and saturations on the actuation system. Based on Proximal Policy Optimization (PPO) and Almost Lyapunov functions in an actor-critic scheme, it allows to robustly learn a controller along with stability certificates, in presence of model uncertainties. Simulation results are proposed to illustrate the approach.
In this work, a leader-follower tracking and formation control strategy for mobile robots (MRs) with uncertain dynamics is proposed. This strategy utilizes a continual lifelong safe reinforcement learning (CLSRL) fram...
详细信息
ISBN:
(纸本)9798350370959;9798350370942
In this work, a leader-follower tracking and formation control strategy for mobile robots (MRs) with uncertain dynamics is proposed. This strategy utilizes a continual lifelong safe reinforcement learning (CLSRL) framework based on multilayer neural networks (MNNs). The proposed design employs actor-critic MNNs, incorporating a barrier function. This function is derived from the Bellman optimality principle. It addresses the state constraints throughout the control design process. A novel online continual lifelong learning (CLL) method is introduced for MR formation. This method leverages the Bellman residual error for weight significance in MNNs. It addresses catastrophic forgetting and interlayer dependence through layer-specific regularizers. Novel weight update laws are proposed. The simulation results show a 35% improvement in performance.
In this paper, we propose a new algorithm for adversarial attacks to direct data-driven control. In direct data-driven control, the controller is directly designed from inputoutput data of the system to be controlled ...
ISBN:
(纸本)9798350370959;9798350370942
In this paper, we propose a new algorithm for adversarial attacks to direct data-driven control. In direct data-driven control, the controller is directly designed from inputoutput data of the system to be controlled without system identification. Previous research has shown that direct datadriven control is vulnerable to adversarial attacks which add small perturbations to the data, and has proposed the directed gradient sign method (DGSM) as a method to generate severe attacks. We introduce a new method, referred to as the iterative directed gradient sign method (IDGSM), based on DGSM and the projected gradient method. Several numerical examples with comparison of IDGSM and DGSM demonstrate that IDGSM generates more sophisticated adversarial attacks than DGSM. Furthermore, we identify the role of the regularization terms through stability analysis using the Lyapunov equation. The regularization term are defense mechanisms against disturbances, and their effectiveness against adversarial attacks has been numerically validated. We provide a theoretical explanation for their effectiveness against attacks.
暂无评论