Learning-based model predictive control on a quadrotor: Onboard implementation and experimental results

Bouffard, Patrick; Aswani, Anil; Tomlin, Claire J.

doi:10.1109/icra.2012.6225035

Cited by 163 publications

(117 citation statements)

References 26 publications

Supporting

Mentioning

117

Contrasting

Order By: Relevance

“…Adaptive formulations seek to estimate the uncertainty in the dynamics and update the predictive model to more accurately anticipate the system's interaction with the constraints [4,8,11,27]. However, in practice, this may still lead to constraint violations due to the difference in timescales between the disturbance estimator and high-frequency noise in the state estimate.…”

Section: Introductionmentioning

confidence: 99%

Experience-driven Predictive Control with Robust Constraint Satisfaction under Time-Varying State Uncertainty

Desaraju

Spitzer

Michael

2017

Robotics: Science and Systems XIII

View full text Add to dashboard Cite

Abstract-We present an extension to Experience-driven Predictive Control (EPC) that leverages a Gaussian belief propagation strategy to compute an uncertainty set bounding the evolution of the system state in the presence of time-varying state uncertainty. This uncertainty set is used to tighten the constraints in the predictive control formulation via a chance constrained approach, thereby providing a probabilistic guarantee of constraint satisfaction. The parameterized form of the controllers produced by EPC coupled with online uncertainty estimates ensures this robust constraint satisfaction property persists even as the system switches controllers and experiences variations in the uncertainty model. We validate the online performance and robust constraint satisfaction of the proposed Robust EPC algorithm through a series of experimental trials with a small quadrotor platform subjected to changes in state estimate quality.

show abstract

Section: Introductionmentioning

confidence: 99%

Experience-driven Predictive Control with Robust Constraint Satisfaction under Time-Varying State Uncertainty

Desaraju

Spitzer

Michael

2017

Robotics: Science and Systems XIII

View full text Add to dashboard Cite

show abstract

“…for a non-switching based MPC [7,8,10]. One approach therefore, could be to continue to learn using a smaller learning-rate r and using estimates of model error only (that is not using the tracking error).…”

Section: Stability Analysismentioning

confidence: 98%

“…Several authors have studied adaptive-MPC architectures that rely on variants of the certainty equivalence principle [7][8][9]. In general, while significant progress has been made, including flight-testing in controlled environments [10] the presence of learning transients prevent a general non-conservative solution to be formed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A concurrent learning adaptive-optimal control architecture for nonlinear systems

Chowdhary

Mühlegg

How

et al. 2013

52nd IEEE Conference on Decision and Control

View full text Add to dashboard Cite

A concurrent learning adaptive-optimal control architecture is presented that combines learning-focused di rect adaptive controllers with model predictive control for guaranteeing safety during adaptation for nonlinear systems. Exponential parameter convergence properties of concurrent learning adaptive controllers are leveraged to learn a feedback linearization signal that reduces a nonlinear system to an approximation of a linear system for which an optimal solution is known or can be easily computed online. Stability of the overall architecture is analyzed, and numerical simulations on a wing-rock dynamics model are presented in presence of significant system uncertainty, parameter variation, and measurement noise.

show abstract

“…In [11] a real time model predictive position control is implemented using sigmoid functions to model obstacles. Finally, in [12] Learning Based MPC is used to catch balls in the air and correct the ground effect.…”

Section: Introductionmentioning

confidence: 99%

Evasive Maneuvering for UAVs: An MPC Approach

Castillo-Lopez

Olivares-Méndez

Voos

2017

ROBOT 2017: Third Iberian Robotics Conference

View full text Add to dashboard Cite

Abstract. Flying autonomously in a workspace populated by obstacles is one of the main goals when working with Unmanned Aerial Vehicles (UAV). To address this challenge, this paper presents a model predictive flight controller that drives the UAV through collision-free trajectories to reach a given pose or follow a waypoint path. The major advantage of this approach lies on the inclusion of threedimensional obstacle avoidance in the control layer by adding ellipsoidal constraints to the optimal control problem. The obstacles can be added, moved and resized online, providing a way to perform waypoint navigation without the need of motion planning. In addition, the delays of the system are cosidered in the prediction by an experimental first order with delay model of the system. Successful experiments in 3D path tracking and obstacle avoidance validates its effectiveness for sense-and-avoid and surveillance applications presenting the proper structure to extent its autonomy and applications.

show abstract

Learning-based model predictive control on a quadrotor: Onboard implementation and experimental results

Cited by 163 publications

References 26 publications

Experience-driven Predictive Control with Robust Constraint Satisfaction under Time-Varying State Uncertainty

Experience-driven Predictive Control with Robust Constraint Satisfaction under Time-Varying State Uncertainty

A concurrent learning adaptive-optimal control architecture for nonlinear systems

Evasive Maneuvering for UAVs: An MPC Approach

Contact Info

Product

Resources

About