Trajectory tracking and obstacle avoidance in robot formations present significant challenges, especially in confined spaces. Traditional methods are constrained by offline optimization problems and are poorly suited ...
详细信息
Trajectory tracking and obstacle avoidance in robot formations present significant challenges, especially in confined spaces. Traditional methods are constrained by offline optimization problems and are poorly suited for addressing these complexities in an online context. This paper presents a safety-critical model predictive control (MPC) strategy for tracking and obstacle avoidance of wheeled mobile robot (WMR) formations. Specifically, to improve the accuracy of obstacle avoidance in WMR formations, a novel discrete-time control barrier function (DCBF)-based polyhedral collision avoidance constraint is integrated into the MPC optimization framework. Then, by leveraging the Lagrangian dual function and strong duality of convex optimization, the implicit DCBF constraints are converted into equivalent explicit forms. This eliminates the nested optimization problem in the MPC framework and significantly reduces computation time. Theoretical analysis confirms that the new DCBF constraints maintain the feasible range of the original constraints. Finally, simulations and hardware experiments are performed on WMR formations to validate the effectiveness of the proposed method.
Safety is a critical component for dynamic systems. Simultaneously imposing multiple dynamic constraints on the dynamic systems for safety adjustment is a challenging problem. In recent years, controlbarrierfunction...
详细信息
ISBN:
(纸本)9789881563804
Safety is a critical component for dynamic systems. Simultaneously imposing multiple dynamic constraints on the dynamic systems for safety adjustment is a challenging problem. In recent years, controlbarrierfunctions (CBFs) have been proposed to deal with the safety problem of dynamic systems. In general, CBFs are combined with control Lyapunov functions (CLFs) by quadratic programming (QP) to construct optimal controller to guarantee the safe trajectory tracking of dynamic systems. In this paper, an augmented CBF is proposed to guarantee the safety of a class of discrete-timecontrol systems. Also, control input perturbation (CIP) algorithm is provided to solve the deadlock problem. Additionally, in order to improve the efficiency of solving QP, switching control is used to reduce the solution time of QP greatly. Finally, the effectiveness of the proposed methods is verified by adjusting the set point of the 3D double integral (3DDI) dynamic system.
In this paper, we propose a safety-critical formation control method based on distributed nonlinear model predictive control strategy, which controls the path following and formation maintenance of the multiple mobile...
详细信息
ISBN:
(纸本)9798350334722
In this paper, we propose a safety-critical formation control method based on distributed nonlinear model predictive control strategy, which controls the path following and formation maintenance of the multiple mobile robots, while ensuring the collision avoidance. Firstly, we adopt the distributed framework with high real time performance. Secondly, based on the distributed optimization framework, discrete-time control barrier function constraints are transformed into smooth differentiable constraints to complete the polytopic obstacle avoidance with a small horizon by using the strong duality of convex optimization. Finally, the simulation results of three robots are given to prove the effectiveness of the proposed algorithm, and it can realize the local path generation based on real-time optimization in the narrow environment.
This paper presents a discrete-time dynamical system model learning method from demonstration while providing probabilistic guarantees on the safety and stability of the learned model. The controlled dynamic model of ...
详细信息
This paper presents a discrete-time dynamical system model learning method from demonstration while providing probabilistic guarantees on the safety and stability of the learned model. The controlled dynamic model of a discrete-time system with a zero-mean Gaussian process noise is approximated using an Extreme Learning Machine (ELM) whose parameters are learned subject to chance constraints derived using a discrete-time control barrier function and discrete-timecontrol Lyapunov function in the presence of the ELM reconstruction error. To estimate the ELM parameters a quadratically constrained quadratic program (QCQP) is developed subject to the constraints that are only required to be evaluated at sampled points. Simulations validate that the system model learned using the proposed method can reproduce the demonstrations inside a prescribed safe set while converging to the desired goal location starting from various different initial conditions inside the safe set. Furthermore, it is shown that the learned model can adapt to changes in goal location during reproductions without violating the stability and safety constraints.
Safety is a critical component for dynamic systems. Simultaneously imposing multiple dynamic constraints on the dynamic systems for safety adjustment is a challenging problem. In recent years, controlbarrierfunction...
详细信息
Safety is a critical component for dynamic systems. Simultaneously imposing multiple dynamic constraints on the dynamic systems for safety adjustment is a challenging problem. In recent years, controlbarrierfunctions（CBFs） have been proposed to deal with the safety problem of dynamic systems. In general, CBFs are combined with control Lyapunov functions（CLFs） by quadratic programming（QP） to construct optimal controller to guarantee the safe trajectory tracking of dynamic systems. In this paper, an augmented CBF is proposed to guarantee the safety of a class of discrete-timecontrol systems. Also,control input perturbation（CIP） algorithm is provided to solve the deadlock problem. Additionally, in order to improve the efficiency of solving QP, switching control is used to reduce the solution time of QP greatly. Finally, the effectiveness of the proposed methods is verified by adjusting the set point of the 3 D double integral（3 DDI） dynamic system.
In this paper,we propose a safety-critical formation control method based on distributed nonlinear model predictive control strategy,which controls the path following and formation maintenance of the multiple mobile r...
详细信息
In this paper,we propose a safety-critical formation control method based on distributed nonlinear model predictive control strategy,which controls the path following and formation maintenance of the multiple mobile robots,while ensuring the collision ***,we adopt the distributed framework with high real time ***,based on the distributed optimization framework,discrete-time control barrier function constraints are transformed into smooth differentiable constraints to complete the polytopic obstacle avoidance with a small horizon by using the strong duality of convex ***,the simulation results of three robots are given to prove the effectiveness of the proposed algorithm,and it can realize the local path generation based on real-time optimization in the narrow environment.
暂无评论