arXiv subjects

Stefano Di Cairano

Publications and source records attributed to Stefano Di Cairano.

33 records · Page 2Linked to original sources

arXiv2023

Real-time Mixed-Integer Quadratic Programming for Vehicle Decision Making and Motion Planning

We develop a real-time feasible mixed-integer programming-based decision making (MIP-DM) system for automated driving. Using a linear vehicle model in a road-aligned coordinate frame, the lane change constraints, collision avoidance and traffic rules can be formulated as mixed-integer inequalities, resulting in a mixed-integer quadratic program (MIQP). The proposed MIP-DM simultaneously performs maneuver selection and trajectory generation by solving the MIQP at each sampling time instant. While solving MIQPs in real time has been considered intractable in the past, we show that our recently developed solver BB-ASIPM is capable of solving MIP-DM problems on embedded hardware in real time. The performance of this approach is illustrated in simulations in various scenarios including merging points and traffic intersections, and hardware-in-the-loop simulations on dSPACE Scalexio and MicroAutoBox-III. Finally, we present results from hardware experiments on small-scale automated vehicles.

math.OC↗

arXiv2023

Robust Parameter Estimation for Hybrid Dynamical Systems

We consider the problem of estimating a vector of unknown constant parameters for a class of hybrid dynamical systems -- that is, systems whose state variables exhibit both continuous (flow) and discrete (jump) evolution. Using a hybrid systems framework, we propose a hybrid estimation algorithm that can operate during both flows and jumps that, under a notion of hybrid persistence of excitation, guarantees convergence of the parameter estimate to the true value. Furthermore, we show that the parameter estimate is input-to-state stable with respect to a class of hybrid disturbances. Simulation results including a spacecraft application show the merits of our proposed approach.

math.OC↗

arXiv2023

Tailored Presolve Techniques in Branch-and-Bound Method for Fast Mixed-Integer Optimal Control Applications

Mixed-integer model predictive control (MI-MPC) can be a powerful tool for modeling hybrid control systems. In case of a linear-quadratic objective in combination with linear or piecewise-linear system dynamics and inequality constraints, MI-MPC needs to solve a mixed-integer quadratic program (MIQP) at each sampling time step. This paper presents a collection of block-sparse presolve techniques to efficiently remove decision variables, and to remove or tighten inequality constraints, tailored to mixed-integer optimal control problems (MIOCP). In addition, we describe a novel heuristic approach based on an iterative presolve algorithm to compute a feasible but possibly suboptimal MIQP solution. We present benchmarking results for a C code implementation of the proposed BB-ASIPM solver, including a branch-and-bound (B&B) method with the proposed tailored presolve techniques and an active-set based interior point method (ASIPM), compared against multiple state-of-the-art MIQP solvers on a case study of motion planning with obstacle avoidance constraints. Finally, we demonstrate the computational performance of the BB-ASIPM solver on the dSPACE Scalexio real-time embedded hardware using a second case study of stabilization for an underactuated cart-pole with soft contacts.

math.OC↗

arXiv2023

Physics-Informed Machine Learning for Modeling and Control of Dynamical Systems

Physics-informed machine learning (PIML) is a set of methods and tools that systematically integrate machine learning (ML) algorithms with physical constraints and abstract mathematical models developed in scientific and engineering domains. As opposed to purely data-driven methods, PIML models can be trained from additional information obtained by enforcing physical laws such as energy and mass conservation. More broadly, PIML models can include abstract properties and conditions such as stability, convexity, or invariance. The basic premise of PIML is that the integration of ML and physics can yield more effective, physically consistent, and data-efficient models. This paper aims to provide a tutorial-like overview of the recent advances in PIML for dynamical system modeling and control. Specifically, the paper covers an overview of the theory, fundamental concepts and methods, tools, and applications on topics of: 1) physics-informed learning for system identification; 2) physics-informed learning for control; 3) analysis and verification of PIML models; and 4) physics-informed digital twins. The paper is concluded with a perspective on open challenges and future research opportunities.

eess.SY↗

arXiv2023

Automated Controller Calibration by Kalman Filtering

This paper proposes a method for calibrating control parameters. Examples of such control parameters are gains of PID controllers, weights of a cost function for optimal control, filter coefficients, the sliding surface of a sliding mode controller, or weights of a neural network. Hence, the proposed method can be applied to a wide range of controllers. The method uses a Kalman filter that estimates control parameters, using data of closed-loop system operation. The control parameter calibration is driven by a training objective, which encompasses specifications on the performance of the dynamical system. The performance-driven calibration method tunes the parameters online and robustly, is computationally efficient, has low data storage requirements, and is easy to implement making it appealing for many real-time applications. Simulation results show that the method is able to learn control parameters quickly, is able to tune the parameters to compensate for disturbances, and is robust to noise. A simulation study with the high-fidelity vehicle simulator CarSim shows that the method can calibrate controllers of a complex dynamical system online, which indicates its applicability to a real-world system. We also verify the real-time feasibility on an embedded platform with automotive-grade processors by implementing our method on a dSPACE MicroAutoBox-II rapid prototyping unit.

eess.SY↗

arXiv2022

Mobility, Communication and Computation Aware Federated Learning for Internet of Vehicles

While privacy concerns entice connected and automated vehicles to incorporate on-board federated learning (FL) solutions, an integrated vehicle-to-everything communication with heterogeneous computation power aware learning platform is urgently necessary to make it a reality. Motivated by this, we propose a novel mobility, communication and computation aware online FL platform that uses on-road vehicles as learning agents. Thanks to the advanced features of modern vehicles, the on-board sensors can collect data as vehicles travel along their trajectories, while the on-board processors can train machine learning models using the collected data. To take the high mobility of vehicles into account, we consider the delay as a learning parameter and restrict it to be less than a tolerable threshold. To satisfy this threshold, the central server accepts partially trained models, the distributed roadside units (a) perform downlink multicast beamforming to minimize global model distribution delay and (b) allocate optimal uplink radio resources to minimize local model offloading delay, and the vehicle agents conduct heterogeneous local model training. Using real-world vehicle trace datasets, we validate our FL solutions. Simulation shows that the proposed integrated FL platform is robust and outperforms baseline models. With reasonable local training episodes, it can effectively satisfy all constraints and deliver near ground truth multi-horizon velocity and vehicle-specific power predictions.

cs.LG↗

arXiv2022

Autonomous Vehicle Parking in Dynamic Environments: An Integrated System with Prediction and Motion Planning

This paper presents an integrated motion planning system for autonomous vehicle (AV) parking in the presence of other moving vehicles. The proposed system includes 1) a hybrid environment predictor that predicts the motions of the surrounding vehicles and 2) a strategic motion planner that reacts to the predictions. The hybrid environment predictor performs short-term predictions via an extended Kalman filter and an adaptive observer. It also combines short-term predictions with a driver behavior cost-map to make long-term predictions. The strategic motion planner comprises 1) a model predictive control-based safety controller for trajectory tracking; 2) a search-based retreating planner for finding an evasion path in an emergency; 3) an optimization-based repairing planner for planning a new path when the original path is invalidated. Simulation validation demonstrates the effectiveness of the proposed method in terms of initial planning, motion prediction, safe tracking, retreating in an emergency, and trajectory repairing.

cs.RO↗

arXiv2020

PRESAS: Block-Structured Preconditioning of Iterative Solvers within a Primal Active-Set Method for fast MPC

Model predictive control (MPC) for linear dynamical systems requires solving an optimal control structured quadratic program (QP) at each sampling instant. This paper proposes a primal active-set strategy (PRESAS) for the efficient solution of such block-sparse QPs, based on a preconditioned iterative solver to compute the search direction in each iteration. Rank-one factorization updates of the preconditioner result in a per-iteration computational complexity of $\mathcal{O}(N m^2)$, where $m$ denotes the number of state and control variables and $N$ the number of control intervals. Three different block-structured preconditioning techniques are presented and their numerical properties are studied further. In addition, an augmented Lagrangian based implementation is proposed to avoid a costly initialization procedure to find a primal feasible starting point. Based on a standalone C code implementation, we illustrate the computational performance of PRESAS against current state of the art QP solvers for multiple linear and nonlinear MPC case studies. We also show that the solver is real-time feasible on a dSPACE MicroAutoBox-II rapid prototyping unit for vehicle control applications, and numerical reliability is illustrated based on experimental results from a testbench of small-scale autonomous vehicles.

math.OC↗

arXiv2019

A Structure Exploiting Branch-and-Bound Algorithm for Mixed-Integer Model Predictive Control

Mixed-integer model predictive control (MI-MPC) requires the solution of a mixed-integer quadratic program (MIQP) at each sampling instant under strict timing constraints, where part of the state and control variables can only assume a discrete set of values. Several applications in automotive, aerospace and hybrid systems are practical examples of how such discrete-valued variables arise. We utilize the sequential nature and the problem structure of MI-MPC in order to provide a branch-and-bound algorithm that can exploit not only the block-sparse optimal control structure of the problem but that can also be warm started by propagating information from branch-and-bound trees and solution paths at previous time steps. We illustrate the computational performance of the proposed algorithm and compare against current state-of-the-art solvers for multiple MPC case studies, based on a preliminary implementation in MATLAB and C code.

math.OC↗

arXiv2017

Path Planning using Positive Invariant Sets

We present an algorithm for steering the output of a linear system from a feasible initial condition to a desired target position, while satisfying input constraints and non-convex output constraints. The system input is generated by a collection of local linear state-feedback controllers. The path-planning algorithm selects the appropriate local controller using a graph search, where the nodes of the graph are the local controllers and the edges of the graph indicate when it is possible to transition from one local controller to another without violating input or output constraints. We present two methods for computing the local controllers. The first uses a fixed-gain controller and scales its positive invariant set to satisfy the input and output constraints. We provide a linear program for determining the scale-factor and a condition for when the linear program has a closed-form solution. The second method designs the local controllers using a semi-definite program that maximizes the volume of the positive invariant set that satisfies state and input constraints. We demonstrate our path-planning algorithm on docking of a spacecraft. The semi-definite programming based control design has better performance but requires more computation.

eess.SY↗

arXiv2016

MPC on manifolds with an application to the control of spacecraft attitude on SO(3)

We develop a model predictive control (MPC) design for systems with discrete-time dynamics evolving on smooth manifolds. We show that the properties of conventional MPC for dynamics evolving on $\mathbb R^n$ are preserved and we develop a design procedure for achieving similar properties. We also demonstrate that for discrete-time dynamics on manifolds with Euler characteristic not equal to 1, there do not exist globally stabilizing, continuous control laws. The MPC law is able to achieve global asymptotic stability on these manifolds, because the MPC law may be discontinuous. We apply the method to spacecraft attitude control, where the spacecraft attitude evolves on the Lie group SO(3) and for which a continuous globally stabilizing control law does not exist. In this case, the MPC law is discontinuous and achieves global stability.

math.OC↗

arXiv2015

ADMM for Convex Quadratic Programs: Q-Linear Convergence and Infeasibility Detection

In this paper, we analyze the convergence of Alternating Direction Method of Multipliers (ADMM) on convex quadratic programs (QPs) with linear equality and bound constraints. The ADMM formulation alternates between an equality constrained QP and a projection on the bounds. Under the assumptions of (i) positive definiteness of the Hessian of the objective projected on the null space of equality constraints (reduced Hessian), and (ii) linear independence constraint qualification holding at the optimal solution, we derive an upper bound on the rate of convergence to the solution at each iteration. In particular, we provide an explicit characterization of the rate of convergence in terms of: (a) the eigenvalues of the reduced Hessian, (b) the cosine of the Friedrichs angle between the subspace spanned by equality constraints and the subspace spanned by the gradients of the components that are active at the solution and (c) the distance of the inactive components of solution from the bounds. Using this analysis we show that if the QP is feasible, the iterates converge at a Q-linear rate and prescribe an optimal setting for the ADMM step-size parameter. For infeasible QPs, we show that the primal variables in ADMM converge to minimizers of the Euclidean distance between the hyperplane defined by the equality constraints and the convex set defined by the bounds. The multipliers for the bound constraints are shown to diverge along the range space of the equality constraints. Using this characterization, we also propose a termination criterion for ADMM. Numerical examples are provided to illustrate the theory through experiments.

math.OC↗

arXiv2015

Indirect-adaptive Model Predictive Control for Linear Systems with Polytopic Uncertainty

We develop an indirect-adaptive model predictive control algorithm for uncertain linear systems subject to constraints. The system is modeled as a polytopic linear parameter varying system where the convex combination vector is constant but unknown. Robust constraint satisfaction is obtained by constraints enforcing a robust control invariant. The terminal cost and set are constructed from a parameter-dependent Lyapunov function and the associated control law. The proposed design ensures robust constraint satisfaction and recursive feasibility, is input-to-state stable with respect to the parameter estimation error and it only requires the online solution of quadratic programs.

eess.SY↗

arXiv2015

Explicit model predictive control accuracy analysis

Model Predictive Control (MPC) can efficiently control constrained systems in real-time applications. MPC feedback law for a linear system with linear inequality constraints can be explicitly computed off-line, which results in an off-line partition of the state space into non-overlapped convex regions, with affine control laws associated to each region of the partition. An actual implementation of this explicit MPC in low cost micro-controllers requires the data to be "quantized", i.e. represented with a small number of memory bits. An aggressive quantization decreases the number of bits and the controller manufacturing costs, and may increase the speed of the controller, but reduces accuracy of the control input computation. We derive upper bounds for the absolute error in the control depending on the number of quantization bits and system parameters. The bounds can be used to determine how many quantization bits are needed in order to guarantee a specific level of accuracy in the control input.

eess.SY↗

arXiv2014

Extremum Seeking-based Iterative Learning Linear MPC

In this work we study the problem of adaptive MPC for linear time-invariant uncertain models. We assume linear models with parametric uncertainties, and propose an iterative multi-variable extremum seeking (MES)-based learning MPC algorithm to learn on-line the uncertain parameters and update the MPC model. We show the effectiveness of this algorithm on a DC servo motor control example.

eess.SY↗