Article

Safe Planning And Control Of Autonomous Systems: Robust Predictive Algorithms

Authors:
To read the full-text of this research, you can request a copy directly from the author.

Abstract

Safe autonomous operation of dynamical systems has become one of the most important research problems. Algorithms for planning and control of such systems are now finding place on production vehicles, and are fast becoming ubiquitous on the roads and air-spaces. However most algorithms for such operations, that provide guarantees, either do not scale well or rely on over-simplifying abstractions that make them impractical for real world implementations. On the other hand, the algorithms that are computationally tractable and amenable to implementation generally lack any guarantees on their behavior. In this work, we aim to bridge the gap between provable and scalable planning and control for dynamical systems. The research covered herein can be broadly categorized into: i) multi-agent planning with temporal logic specifications, and ii) robust predictive control that takes into account the performance of the perception algorithms used to process information for control. In the first part, we focus on multi-robot systems with complicated mission requirements, and develop a planning algorithm that can take into account a) spatial, b) temporal and c) reactive mission requirements across multiple robots. The algorithm not only guarantees continuous time satisfaction of the mission requirements, but also that the generated trajectories can be followed by the robot. The other part develops a robust, predictive control algorithm to control the the dynamical system to follow the trajectories generated by the first part, within some desired bounds. This relies on a contract-based framework wherein the control algorithm controls the dynamical system as well as a resource/quality trade-off in a perception-based state estimation algorithm. We show that this predictive algorithm remains feasible with respect to constraints while following a desired trajectory, and also stabilizes the dynamical system under control. Through simulations, as well as experiments on actual robotic systems, we show that the planning method is computationally efficient as well as scales better than other state-of-the art algorithms that use similar formal specification. We also show that the robust control algorithm provides better control performance, and is also computationally more efficient than similar algorithms that do not leverage the resource/quality trade-off of the perception-based state estimator

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the author.

ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
The correct and timely completion of the sensing and action loop is of utmost importance in safety critical autonomous systems. A crucial part of the performance of this feedback control loop are the computation time and accuracy of the estimator which produces state estimates used by the controller. These state estimators, especially those used for localization, often use computationally expensive perception algorithms like visual object tracking. With on-board computers on autonomous robots being computationally limited, the computation time of a perception-based estimation algorithm can at times be high enough to result in poor control performance. In this work, we develop a framework for co-design of anytime estimation and robust control algorithms while taking into account computation delays and estimation inaccuracies. This is achieved by constructing a perception-based anytime estimator from an off-the-shelf perception-based estimation algorithm, and in the process we obtain a trade-off curve for its computation time versus estimation error. This information is used in the design of a robust predictive control algorithm that at run-time decides a contract for the estimator, or the mode of operation of estimator, in addition to trying to achieve its control objectives at a reduced computation energy cost. In cases where the estimation delay can result in possibly degraded control performance, we provide an optimal manner in which the controller can use this trade-off curve to reduce estimation delay at the cost of higher inaccuracy, all the while guaranteeing that control objectives are robustly satisfied. Through experiments on a hexrotor platform running a visual odometry algorithm for state estimation, we show how our method results in upto a 10% improvement in control performance while saving 5-6% in computation energy as compared to a method that does not leverage the co-design.
Conference Paper
Full-text available
We present a perception-aware model predictive control framework for quadrotors that unifies control and planning with respect to action and perception objectives. Our framework leverages numerical optimization to compute trajectories that satisfy the system dynamics and require control inputs within the limits of the platform. Simultaneously, it optimizes perception objectives for robust and reliable sens- ing by maximizing the visibility of a point of interest and minimizing its velocity in the image plane. Considering both perception and action objectives for motion planning and control is challenging due to the possible conflicts arising from their respective requirements. For example, for a quadrotor to track a reference trajectory, it needs to rotate to align its thrust with the direction of the desired acceleration. However, the perception objective might require to minimize such rotation to maximize the visibility of a point of interest. A model-based optimization framework, able to consider both perception and action objectives and couple them through the system dynamics, is therefore necessary. Our perception-aware model predictive control framework works in a receding-horizon fashion by iteratively solving a non-linear optimization problem. It is capable of running in real-time, fully onboard our lightweight, small-scale quadrotor using a low-power ARM computer, to- gether with a visual-inertial odometry pipeline. We validate our approach in experiments demonstrating (I) the contradiction between perception and action objectives, and (II) improved behavior in extremely challenging lighting conditions.
Conference Paper
Full-text available
The problem of safe planning and control for multi-agent systems across a variety of missions is of critical importance , as the scope of tasks assigned to such systems increases. In this paper, we present an approach to solve this problem for multi-quadrotor missions. Given a mission expressed in Signal Temporal Logic (STL), our controller maximizes robustness to generate trajectories for the quadrotors that satisfy the STL specification in continuous-time. We also show that the constraints on our optimization guarantees that these trajectories can be tracked nearly perfectly by lower level off-the-shelf position and attitude controllers. Our approach avoids the oversimplifying abstractions found in many planning methods, while retaining the expressiveness of missions encoded in STL allowing us to handle complex spatial, temporal and reactive requirements. Through experiments, both in simulation and on actual quadrotors, we show the performance, scalability and real-time applicability of our method.
Article
Full-text available
This paper proposes a sampling-based algorithm for optimal multi-robot control synthesis under global Linear Temporal Logic formulas. Existing planning approaches under global temporal goals rely on graph search techniques applied to a product automaton constructed among the robots. In this paper, we propose a new sampling-based algorithm that builds incrementally trees that approximate the state-space and transitions of the synchronous product automaton. By approximating the product automaton by a tree rather than representing it explicitly, we require much fewer resources to store it and motion plans can be found by tracing sequences of parent nodes without the need for sophisticated graph search techniques. This significantly increases scalability of our algorithm compared to existing optimal control synthesis methods. We also show that the proposed algorithm is probabilistically complete and asymptotically optimal. Finally, we present numerical experiments that show that our sampling-based approach can synthesize optimal plans from product automata with billions of states, which is not possible using standard optimal control synthesis algorithms or off-the-shelf model checkers.
Conference Paper
Full-text available
Decisions on how to best optimize energy systems operations are becoming ever so complex and conflicting, that model-based predictive control (MPC) algorithms must play an important role. However, a key factor prohibiting the widespread adoption of MPC in buildings, is the cost, time, and effort associated with learning first-principles based dynamical models of the underlying physical system. This paper introduces an alternative approach for implementing finite-time receding horizon control using control-oriented data-driven models. We call this approach Data Predictive Control (DPC). Specifically, by utilizing separation of variables, two novel algorithms for implementing DPC using a single regression tree and with regression trees ensembles (random forest) are presented. The data predictive controller enables the building operator to trade off energy consumption against thermal comfort without having to learn white/grey box models of the systems dynamics. We present a comprehensive numerical study which compares the performance of DPC with an MPC based energy management strategy, using a single zone building model. Our results demonstrate that performance of DPC is comparable to an MPC controller, with only 3.8% additional cost in terms of optimal objective function and within 95% in terms of R2 score, thereby making it an alluring alternative to MPC, whenever the associated cost of learning the model is high.
Conference Paper
Full-text available
Multi-Agent Path Finding (MAPF) is well studied in both AI and robotics. Given a discretized environment and agents with assigned start and goal locations, MAPF solvers from AI find collision-free paths for hundreds of agents with user-provided sub-optimality guarantees. However, they ignore that actual robots are subject to kinematic constraints (such as finite maximum velocity limits) and suffer from imperfect plan-execution capabilities. We therefore introduce MAPF-POST, a novel approach that makes use of a simple temporal network to postprocess the output of a MAPF solver in polynomial time to create a plan-execution schedule that can be executed on robots. This schedule works on non-holonomic robots, takes their maximum translational and rotational velocities into account, provides a guaranteed safety distance between them, and exploits slack to absorb imperfect plan executions and avoid time-intensive replanning in many cases. We evaluate MAPF-POST in simulation and on differential-drive robots, showcasing the practicality of our approach.
Article
Full-text available
Model predictive control problems for constrained hybrid systems are usually cast as mixed-integer optimization problems (MIP). However, commercial MIP solvers are designed to run on desktop computing platforms and are not suited for embedded applications which are typically restricted by limited computational power and memory. To alleviate these restrictions, we develop a novel low-complexity, iterative method for a class of non-convex, non-smooth optimization problems. This class of problems encompasses hybrid model predictive control problems where the dynamics are piece-wise affine (PWA). We give conditions such that the proposed algorithm has fixed points and show that, under practical assumptions, our method is guaranteed to converge locally to local minima. This is in contrast to other low-complexity methods in the literature, such as the non-convex alternating directions method of multipliers (ADMM), for which no such guarantees are known for this class of problems. By interpreting the PWA dynamics as a union of polyhedra we can exploit the problem structure and develop an algorithm based on operator splitting procedures. Our algorithm departs from the traditional MIP formulation, and leads to a simple, embeddable method that only requires matrix-vector multiplications and small-scale projections onto polyhedra. We illustrate the efficacy of the method on two numerical examples, achieving good closed-loop performance with computational times several orders of magnitude smaller compared to state-of-the-art MIP solvers. Moreover, it is competitive with ADMM in terms of suboptimality and computation time, but additionally provides local optimality and local convergence guarantees.
Article
Full-text available
Robot 3D (three-dimension) path planning targets for finding an optimal and collision-free path in a 3D workspace while taking into account kinematic constraints (including geometric, physical, and temporal constraints). The purpose of path planning, unlike motion planning which must be taken into consideration of dynamics, is to find a kinematically optimal path with the least time as well as model the environment completely. We discuss the fundamentals of these most successful robot 3D path planning algorithms which have been developed in recent years and concentrate on universally applicable algorithms which can be implemented in aerial robots, ground robots, and underwater robots. This paper classifies all the methods into five categories based on their exploring mechanisms and proposes a category, called multifusion based algorithms. For all these algorithms, they are analyzed from a time efficiency and implementable area perspective. Furthermore a comprehensive applicable analysis for each kind of method is presented after considering their merits and weaknesses.
Conference Paper
Full-text available
Control software of autonomous robots has stringent real-time requirements that must be met to achieve the control objectives. One source of variability in the performance of a control system is the execution time and accuracy of the state estimator that provides the controller with state information. This estimator is typically perception-based (e.g., Computer Vision-based) and is computationally expensive. When the computational resources of the hardware platform become overloaded, the estimation delay can compromise control performance and even stability. In this paper, we define a framework for co-designing anytime estimation and control algorithms, in a manner that accounts for implementation issues like delays and inaccuracies. We construct an anytime perception-based estimator from standard off-the-shelf Computer Vision algorithms, and show how to obtain a trade-off curve for its delay vs estimate error behavior. We use this anytime estimator in a controller that can use this trade-off curve at runtime to achieve its control objectives at a reduced energy cost. When the estimation delay is too large for correct operation, we provide an optimal manner in which the controller can use this curve to reduce estimation delay at the cost of higher inaccuracy, all the while guaranteeing basic objectives are met. We illustrate our approach on an autonomous hexrotor and demonstrate its advantage over a system that does not exploit co-design.
Article
Full-text available
The Simplex Architecture ensures the safe use of an unverifiable complex controller by using a verified safety controller and verified switching logic. This architecture enables the safe use of high-performance, untrusted, and complex control algorithms without requiring them to be formally verified. Simplex incorporates a supervisory controller and safety controller that will take over control if the unverified logic misbehaves. The supervisory controller should (1) guarantee the system never enters and unsafe state (safety), but (2) use the complex controller as much as possible (minimize conservatism). The problem of precisely and correctly defining this switching logic has previously been considered either using a control-theoretic optimization approach, or through an offline hybrid systems reach ability computation. In this work, we prove that a combined online/offline approach, which uses aspects of the two earlier methods along with a real-time reach ability computation, also maintains safety, but with significantly less conservatism. We demonstrate the advantages of this unified approach on a saturated inverted pendulum system, where the usable region of attraction is 227% larger than the earlier approach.
Article
Full-text available
The theories underlying control engineering and real-time systems engineering use idealized models that mutually abstract from central aspects of the other discipline. Control theory usually assumes jitter-free sampling and negligible (constant) input-output latencies, disregarding complex real-world timing effects. Real-time systems theory uses abstract performance models that neglect the functional behavior and derives worst-case situations with limited expressiveness for control functions, e.g., In physically dominated automotive systems. In this paper, we propose an approach that integrates state-of-the art timing models into functional analysis. We combine physical, control and timing models by representing them as a network of hybrid automata. Closed-loop properties can then be verified on this hybrid automata network by using standard model checkers for hybrid systems. Since the computational complexity is critical for model checking, we discuss abstract models of timing behavior that seem particularly suited for this type of analysis. The approach facilitates systematic co-engineering between both control and real-time disciplines, increasing design efficiency and confidence in the system. The approach is illustrated by analyzing an industrial example, the control software of an electro-mechanical braking system, with the hybrid model checker Space Ex.
Article
Full-text available
This paper presents ORB-SLAM, a feature-based monocular SLAM system that operates in real time, in small and large, indoor and outdoor environments. The system is robust to severe motion clutter, allows wide baseline loop closing and relocalization, and includes full automatic initialization. Building on excellent algorithms of recent years, we designed from scratch a novel system that uses the same features for all SLAM tasks: tracking, mapping, relocalization, and loop closing. A survival of the fittest strategy that selects the points and keyframes of the reconstruction leads to excellent robustness and generates a compact and trackable map that only grows if the scene content changes, allowing lifelong operation. We present an exhaustive evaluation in 27 sequences from the most popular datasets. ORB-SLAM achieves unprecedented performance with respect to other state-of-the-art monocular SLAM approaches. For the benefit of the community, we make the source code public.
Article
Full-text available
Many continuous wavelets are defined in the frequency domain and do not have analytical expressions in the time domain. Meyer wavelet is ordinarily defined in this way. In this note, we derive new straightforward analytical expressions for both the wavelet and scale function for the Meyer basis. The validity of these expressions is corroborated by numerical computations, yielding no approximation error.
Conference Paper
Full-text available
Metric Temporal Logic (MTL) specifications can capture complex state and timing requirements. Given a nonlinear dynamical system and an MTL specification for that system, our goal is to find a trajectory that violates or satisfies the specification. This trajectory can be used as a concrete feedback to the system designer in the case of violation or as a trajectory to be tracked in the case of satisfaction. The search for such a trajectory is conducted over the space of initial conditions, system parameters and input signals. We convert the trajectory search problem into an optimization problem through MTL robust semantics. Robustness quantifies how close the trajectory is to violating or satisfying a specification. Starting from some arbitrary initial condition and parameter and given an input signal, we compute a descent direction in the search space, which leads to a trajectory that optimizes the MTL robustness. This process can be iterated to reach local optima (min or max). We demonstrate the method on examples from the literature.
Chapter
Full-text available
Fuzzy control theory is an emerging area of research. At the core of many engineering problems is the problem of control of different systems. These systems range all the way from classical inverted pendulum to auto-focusing system of a digital camera. Fuzzy control systems have demonstrated their enhanced performance in all these areas. Progress in this domain is very fast and there was critical need of a book that captures all the recent advances both in theory and in applications. Serving this purpose, this book is conceived. This book will provide you a very clear picture of current status of fuzzy control research. This book is intended for researchers, engineers, and postgraduate students specializing in fuzzy systems, control engineering, and robotics.
Conference Paper
Full-text available
In this paper, we provide a Dynamic Programming algorithm for on-line monitoring of the state robustness of Metric Temporal Logic specifications with past time operators. We compute the robustness of MTL with unbounded past and bounded future temporal operators MTL over sampled traces of Cyber-Physical Systems. We implemented our tool in Matlab as a Simulink block that can be used in any Simulink model. We experimentally demonstrate that the overhead of the MTL robustness monitoring is acceptable for certain classes of practical specifications.
Conference Paper
Full-text available
We propose a semi-direct monocular visual odom-etry algorithm that is precise, robust, and faster than current state-of-the-art methods. The semi-direct approach eliminates the need of costly feature extraction and robust matching techniques for motion estimation. Our algorithm operates directly on pixel intensities, which results in subpixel precision at high frame-rates. A probabilistic mapping method that explicitly models outlier measurements is used to estimate 3D points, which results in fewer outliers and more reliable points. Precise and high frame-rate motion estimation brings increased robustness in scenes of little, repetitive, and high-frequency texture. The algorithm is applied to micro-aerial-vehicle state-estimation in GPS-denied environments and runs at 55 frames per second on the onboard embedded computer and at more than 300 frames per second on a consumer laptop. We call our approach SVO (Semi-direct Visual Odometry) and release our implementation as open-source software.
Article
Full-text available
We present a Monte-Carlo optimization technique for finding system behaviors that falsify a metric temporal logic (MTL) property. Our approach performs a random walk over the space of system inputs guided by a robustness metric defined by the MTL property. Robustness is guiding the search for a falsifying behavior by exploring trajectories with smaller robustness values. The resulting testing framework can be applied to a wide class of cyber-physical systems (CPS). We show through experiments on complex system models that using our framework can help automatically falsify properties with more consistency as compared to other means, such as uniform sampling.
Conference Paper
Full-text available
We present a case study on the use of robustness-guided and statistical model checking approaches for simulating risks due to insulin infusion pump usage by diabetic patients. Insulin infusion pumps allow for a continuous delivery of insulin with varying rates and delivery profiles to help patients self-regulate their blood glucose levels. However, the use of infusion pumps and continuous glucose monitors can pose risks to the patient including chronically elevated blood glucose levels (hyperglycemia) or dangerously low glucose levels (hypoglycemia). In this paper, we use mathematical models of the basic insulin-glucose regulatory system in a diabetic patient, insulin infusion pumps, and the user’s interaction with these pumps defined by commonly used insulin infusion strategies for maintaining normal glucose levels. These strategies include common guidelines taught to patients by physicians and certified diabetes educators and have been implemented in commercially available insulin bolus calculators. Furthermore, we model the failures in the devices themselves along with common errors in the usage of the pump. We compose these models together and analyze them using two related techniques: (a) robustness guided state-space search to explore worst-case scenarios and (b) statistical model checking techniques to assess the probabilities of hyper- and hypoglycemia risks. Our technique can be used to identify the worst-case effects of the combination of many different kinds of failures and place high confidence bounds on their probabilities.
Conference Paper
Full-text available
The Multi-Parametric Toolbox is a collection of algorithms for modeling, control, analysis, and deployment of constrained optimal controllers developed under Matlab. It features a powerful geometric library that extends the application of the toolbox beyond optimal control to various problems arising in computational geometry. The new version 3.0 is a complete rewrite of the original toolbox with a more flexible structure that offers faster integration of new algorithms. The numerical side of the toolbox has been improved by adding interfaces to state of the art solvers and by incorporation of a new parametric solver that relies on solving linear-complementarity problems. The toolbox provides algorithms for design and implementation of real-time model predictive controllers that have been extensively tested.
Conference Paper
Full-text available
S-TALIRO is a software toolbox that performs stochastic search for system trajectories that falsify realtime temporal logic specifications. S-TaLiRo is founded on the notion of robustness of temporal logic specifications. In this paper, we present a dynamic programming algorithm for computing the robustness of temporal logic specifications with respect to system trajectories. We also demonstrate that typical automotive functional requirements can be captured and falsified using temporal logics and S-TALIRO.
Conference Paper
Full-text available
The automatic analysis of transient properties of nonlinear dynamical systems is a challenging problem. The problem is even more challenging when complex state-space and timing requirements must be satisfied by the system. Such complex requirements can be captured by Metric Temporal Logic (MTL) specifications. The problem of finding system behaviors that do not satisfy an MTL specification is referred to as MTL falsification. This paper presents an approach for improving stochastic MTL falsification methods by performing local search in the set of initial conditions. In particular, MTL robustness quantifies how correct or wrong is a system trajectory with respect to an MTL specification. Positive values indicate satisfaction of the property while negative values indicate falsification. A stochastic falsification method attempts to minimize the system's robustness with respect to the MTL property. Given some arbitrary initial state, this paper presents a method to compute a descent direction in the set of initial conditions, such that the new system trajectory gets closer to the unsafe set of behaviors. This technique can be iterated in order to converge to a local minimum of the robustness landscape. The paper demonstrates the applicability of the method on some challenging nonlinear systems from the literature.
Article
Full-text available
We present two related anytime algorithms for control of nonlinear systems when the processing resources available are time-varying. The basic idea is to calculate tentative control input sequences for as many time steps into the future as allowed by the available processing resources at every time step. This serves to compensate for the time steps when the processor is not available to perform any control calculations. Using a stochastic Lyapunov function-based approach, we analyze the stability of the resulting closed-loop system for the cases when the processor availability can be modeled as an independent and identically distributed sequence and via an underlying Markov chain. Numerical simulations indicate that the increase in performance due to the proposed algorithms can be significant.
Article
Full-text available
Motion planning is a fundamental research area in robotics. Sampling-based methods offer an efficient solution for what is otherwise a rather challenging dilemma of path planning. Consequently, these methods have been extended further away from basic robot planning into further difficult scenarios and diverse applications. A comprehensive survey of the growing body of work in sampling-based planning is given here. Simulations are executed to evaluate some of the proposed planners and highlight some of the implementation details that are often left unspecified. An emphasis is placed on contemporary research directions in this field. We address planners that tackle current issues in robotics. For instance, real-life kinodynamic planning, optimal planning, replanning in dynamic environments, and planning under uncertainty are discussed. The aim of this paper is to survey the state of the art in motion planning and to assess selected planners, examine implementation details and above all shed a light on the current challenges in motion planning and the promising approaches that will potentially overcome those problems.
Article
This paper describes a holistic method for automatically synthesizing controllers for a team of robots operating in an environment shared with other agents. The proposed approach builds on recent advances in Reactive Mission Planning using Linear Temporal Logic, and Local Motion Planning using convex optimization. A local planner enforces the dynamic constraints of the robot and guarantees collision avoidance in 2D and 3D workspaces. A reactive mission planner takes a high-level specification that captures complex motion sequencing, and generates a correct-by-construction controller guaranteed to achieve the specified behavior and be reactive to sensor events. If there is no controller that fulfills the specification because of possible deadlock in the local planner, a minimal set of human-readable assumptions is generated as a certificate of the conditions on deadlock where the task is guaranteed. This is truly a synergistic method: the low-level motion planner enables scalability of the high-level plan synthesis with respect to dynamic obstacles, and the high-level mission planner enforces correctness of the low-level motion. We provide formal guarantees for our approach and demonstrate it via physical experiments with ground robots and simulations with a team of quadrotors.
Conference Paper
High-fidelity models of airport and airspace capacity enable researchers to study modernization strategies that optimize capacity. The design and the development of airport and airspace capacity models require volumes of detailed aircraft movement data in the terminal airspace. This data, while it exists in the public realm, is highly challenging and cumbersome to collect in large quantities. In this study, we present a methodology to develop our titled Approach Airspace Characterization (AAC) database, fed by flight data scraping techniques and transcriptions of Air Traffic Control voice commands in the terminal airspace. We illustrate the mathematical mechanisms required to assign aircraft movements to specific arrival fixes to develop high-fidelity models of movements. We then present innovative new ways of measuring delay and refining arrival airspace models based on this fine-grained data.
Book
For the first time, a textbook that brings together classical predictive control with treatment of up-to-date robust and stochastic techniques. Model Predictive Control describes the development of tractable algorithms for uncertain, stochastic, constrained systems. The starting point is classical predictive control and the appropriate formulation of performance objectives and constraints to provide guarantees of closed-loop stability and performance. Moving on to robust predictive control, the text explains how similar guarantees may be obtained for cases in which the model describing the system dynamics is subject to additive disturbances and parametric uncertainties. Open- and closed-loop optimization are considered and the state of the art in computationally tractable methods based on uncertainty tubes presented for systems with additive model uncertainty. Finally, the tube framework is also applied to model predictive control problems involving hard or probabilistic constraints for the cases of multiplicative and stochastic model uncertainty. The book provides: • extensive use of illustrative examples; • sample problems; and • discussion of novel control applications such as resource allocation for sustainable development and turbine-blade control for maximized power capture with simultaneously reduced risk of turbulence-induced damage. Graduate students pursuing courses in model predictive control or more generally in advanced or process control and senior undergraduates in need of a specialized treatment will find Model Predictive Control an invaluable guide to the state of the art in this important subject. For the instructor it provides an authoritative resource for the construction of courses.
Article
Cyber-Physical Systems must withstand a wide range of errors, from bugs in their software to attacks on their physical sensors. Given a formal specification of their desired behavior in Metric Temporal Logic (MTL), the robust semantics of the specification provides a notion of system robustness that can be calculated directly on the output behavior of the system, without explicit reference to the various sources or models of the errors. The robustness of the MTL specification has been used both to verify the system offline (via robustness minimization) and to control the system online (to maximize its robustness over some horizon). Unfortunately, the robustness objective function is difficult to work with: it is recursively defined, non-convex and non-differentiable. In this paper, we propose smooth approximations of the robustness. Such approximations are differentiable, thus enabling us to use powerful off-the- shelf gradient descent algorithms for optimizing it. By using them we can also offer guarantees on the performance of the optimization in terms of convergence to minima. We show that the approximation error is bounded to any desired level, and that the approximation can be tuned to the specification. We demonstrate the use of the smooth robustness to control two quad-rotors in an autonomous air traffic control scenario, and for temperature control of a building for comfort.
Conference Paper
Distributed mobile robotics (DMR) involves teams of networked robots navigating in a physical space to achieve tasks in a coordinated fashion. A major challenge in DMR is to program the ensemble of robots with formal guarantees and high assurance of correct operation. To this end, we introduce Drona, a framework for building reliable DMR applications. This paper makes three central contributions: (1) We present a novel and provably correct decentralized asynchronous motion planner that can perform on-the-fly collision-free planning for dynamically generated tasks. Moreover, the motion planner is the first to take into account the fact that distributed robots may have clocks that are only synchronized up to a tolerance, i.e., they are almost synchronous; (2) We formalize the DMR system as a mixed-synchronous system, and present a sound abstraction-based verification approach for DMR systems, and (3) Drona provides a state-machine based language for safe event-driven programming of a DMR system and the code generated by the compiler can be executed on platforms such as the robot operating system (ROS). To demonstrate the efficacy of Drona, we build and verify a priority mail delivery system. Using our abstraction-based verification approach we were able to find, within a few minutes, bugs which could not be found by performing random simulation for several hours. Our verified decentralized motion-planner scales efficiently for large number of robots (upto 128 robots) and workspace sizes (upto a 256×256 grid).
Conference Paper
This paper presents a decentralized multi-agent motion planning method for aerial robots moving among 3D polygonal obstacles. The algorithm combines a prioritized A∗ algorithm for high-level (global) planning along with a barrier functions-based method for low-level (local) coordination and control. We first extend the barriers function method developed in earlier work to treat arbitrary polygonal obstacles. We then combine the prioritized A∗ algorithm to compute waypoints and paths that facilitate the performance of the barrier-based coordination and collision avoidance. We assume that the obstacles are known to the agents, and that each agent knows the state of other agents lying in its sensing area. Simulation and experimental results with quadrotors in 2D and 3D environments demonstrate the efficacy of the proposed approach.
Article
We present a primal-dual interior-point algorithm with a lter line-search method for non- linear programming. Local and global convergence properties of this method were analyzed in previous work. Here we provide a comprehensive description of the algorithm, including the fea- sibility restoration phase for the lter method, second-order corrections, and inertia correction of the KKT matrix. Heuristics are also considered that allow faster performance. This method has been implemented in the IPOPT code, which we demonstrate in a detailed numerical study based on 954 problems from the CUTEr test set. An evaluation is made of several line-search options, and a comparison is provided with two state-of-the-art interior-point codes for nonlin- ear programming.
Conference Paper
A successful method for model predictive control of constrained linear systems uses a local linear control law that, in the presence of disturbances, maintains the disturbed trajectory of the controlled system in a tube whose centre is the nominal trajectory (generated ignoring the disturbance) and whose 'cross-section' is a robust, positively invariant set; robust exponential stability of an invariant set centred on the origin may be established. The purpose of this paper is to show how this successful procedure may be extended to provide robust model predictive control of constrained nonlinear systems. An ancillary problem is proposed, the solution of which provides a local nonlinear control law. The disturbed trajectories lie in a tube and this provides the means for constructing a tube-based robust nonlinear model predictive controller.
Conference Paper
A nonlinear model predictive control scheme guaranteeing robust constraint satisfaction is presented. The scheme is applicable to polynomial or rational systems and guarantees that state, terminal, and output constraints are robustly satisfied despite uncertain and bounded disturbances, parameters, and state measurements or estimates. In addition, for a suitably chosen terminal set, feasibility of the underlying optimization problem at a time instance guarantees that the constraints are robustly satisfied for all future time instances. The proposed scheme utilizes a semi-infinite optimization problem reformulated as a bilevel optimization problem: The outer program determines an input minimizing a performance index for a nominal nonlinear system, while several inner programs certify robust constraint satisfaction. We use convex relaxations to deal with the nonlinear dynamics in the inner programs efficiently. A simulation example is presented to demonstrate the approach.
Conference Paper
Emerging high-performance architectures are anticipated to contain unreliable components that may exhibit soft errors, which silently corrupt the results of computations. Full detection and masking of soft errors is challenging, expensive, and, for some applications, unnecessary. For example, approximate computing applications (such as multimedia processing, machine learning, and big data analytics) can often naturally tolerate soft errors. We present Rely a programming language that enables developers to reason about the quantitative reliability of an application -- namely, the probability that it produces the correct result when executed on unreliable hardware. Rely allows developers to specify the reliability requirements for each value that a function produces. We present a static quantitative reliability analysis that verifies quantitative requirements on the reliability of an application, enabling a developer to perform sound and verified reliability engineering. The analysis takes a Rely program with a reliability specification and a hardware specification that characterizes the reliability of the underlying hardware components and verifies that the program satisfies its reliability specification when executed on the underlying unreliable hardware platform. We demonstrate the application of quantitative reliability analysis on six computations implemented in Rely.
Conference Paper
The guarantee of feasibility given feasibility at initial time is an issue that has been overlooked by many of the recent papers on stochastic model predictive control. Effective solutions have recently been proposed, but these carry considerable online computational load and a degree of conservativism. For the case that the elements of the random additive disturbance vector are independent, the current paper ensures that probabilistic constraints are met and that a quadratic stability condition is satisfied. A numerical example illustrates the efficacy of the proposed algorithm, which achieves tight satisfaction of constraints and thereby attains near-optimal performance.
Conference Paper
Path planning in dynamic environments is signif- icantly more difficult than navigation in static spaces due to the increased dimensionality of the problem, as well as the importance of returning good paths under time constraints. Anytime planners are ideal for these types of problems as they find an initial solution quickly and then improve it as time allows. In this paper, we develop an anytime planner that builds off of Safe Interval Path Planning (SIPP), which is a fast A*-variant for planning in dynamic environments that uses intervals instead of timesteps to represent the time dimension of the problem. In addition, we introduce an optional time- horizon after which the planner drops time as a dimension. On the theoretical side, we show that in the absence of time- horizon our planner can provide guarantees on completeness as well as bounds on the sub-optimality of the solution with respect to the original space-time graph. We also provide simulation experiments for planning for a UAV among 50 dynamic obstacles, where we can provide safe paths for the next 15 seconds of execution within 0.05 seconds. Our results provide a strong evidence for our planner working under real- time constraints.
Article
CVXGEN is a software tool that takes a high level description of a convex optimization problem family, and automatically generates custom C code that compiles into a reliable, high speed solver for the problem family. The current implementation targets problem families that can be transformed, using disciplined convex programming techniques, to convex quadratic programs of modest size. CVXGEN generates simple, flat, library-free code suitable for embedding in real-time applications. The generated code is almost branch free, and so has highly predictable run-time behavior. The combination of regularization (both static and dynamic) and iterative refinement in the search direction computation yields reliable performance, even with poor quality data. In this paper we describe how CVXGEN is implemented, and give some results on the speed and reliability of the automatically generated solvers.
Article
Large variations in the execution times of algorithms characterize many cyber-physical systems (CPS). For example, variations arise in the case of visual object-tracking tasks, whose execution times depend on the contents of the current field of view of the camera. In this paper, we study such a scenario in a small Unmanned Aerial Vehicle (UAV) system with a camera that must detect objects in a variety of conditions ranging from the simple to the complex. Given resource, weight and size constraints, such cyber-physical systems do not have the resources to satisfy the hard-real-time requirements of safe flight along with the need to process highly variable workloads at the highest quality and resolution levels. Hence, tradeoffs have to be made in real-time across multiple levels of criticality of running tasks and their operating points. Specifically, the utility derived from tracking an increasing number of objects may saturate when the mission software can no longer perform the required processing on each individual object. In this paper, we evaluate a new approach called ZS-QRAM (Zero-Slack QoS-based Resource Allocation Model) that maximizes the UAV system utility by explicitly taking into account the diminishing returns on tracking an increasing number of objects. We perform a detailed evaluation of our approach on our UAV system to clearly demonstrate its benefits.