Motion-planning is an essential component of autonomous aerial and terrestrial vehicles. The canonical Motion-planning problem, which is widely studied in the literature, is of planning point-to-point motion while avoiding obstacles. However, the desired degree of vehicular autonomy has steadily risen, and has consequently led to motion-planning problems where a vehicle is required to accomplish a high-level intelligent task, rather than simply move between two points. One way of specifying such intelligent tasks is via linear temporal logic (LTL) formulae. LTL is a formal logic system that includes temporal operators such as always, eventually, and until besides the usual logical operators. For autonomous vehicles, LTL formulae can concisely express tasks such as persistent surveillance, safety requirements, and temporal orders of visits to multiple locations. Recent control theoretic literature has discussed the generation of reference trajectories and/or the synthesis of feedback control laws to enable a vehicle to move in manners that satisfy LTL specifications. A crucial step in such synthesis is the generation of a so-called discrete abstraction of a vehicle kinematic/dynamic model. Typical techniques of generating a discrete abstraction require strong assumptions on controllability and/or linearity. This dissertation discusses fast motion-planning and control techniques to satisfy LTL specifications for vehicle models with nonholonomic kinematic constraints, which do not satisfy the aforesaid assumptions. The main contributions of this dissertation are as follows. First, we present a new technique for constructing discrete abstractions of a Dubins vehicle model (namely, a vehicle that moves forward at a constant speed with a minimum turning radius). This technique relies on the so-called method of lifted graphs and precomputed reachable set calculations. Using this technique, we provide an algorithm to generate vehicle reference trajectories satisfying LTL specifications without requiring complete controllability in the presence of workspace constraints, and without requiring linearity or linearization of the vehicle model. Second, we present a technique for centralized motion-planning for a team of vehicles to collaboratively satisfy a common LTL specification. This technique is also based on the method of lifted graphs. Third, we present an incremental version of the proposed motion-planning techniques, which has an “anytime" property. This property means that a feasible solution is computed quickly, and the iterative updates are made to this solution with a guarantee of convergence to an optimal solution. This version is suited for real-time implementation, where a hard bound on the computation time is imposed. Finally, we present a randomized sampling-based technique for generating reference trajectories that satisfy given LTL specifications. This technique is an alternative to the aforesaid technique based on lifted graphs. We illustrate the proposed techniques using numerical simulation examples. We demonstrate the superiority of the proposed techniques in comparison to the existing literature in terms of computational time and memory requirements.

Autonomous robots like household service robots, self-driving cars and dronesare emerging as important parts of our daily lives in the near future. They need tocomprehend and fulfill complex tasks specified by the users with minimal humanintervention. Also they should be able to handle un-modeled changes and contingentevents in the workspace. More importantly, they shall communicate and collaboratewith each other in an efficient and correct manner. In this thesis, we address theseissues by focusing on the distributed and hybrid control of multi-robot systemsunder complex individual tasks. We start from the nominal case where a single dynamical robot is deployed in astatic and fully-known workspace. Its local tasks are specified as Linear TemporalLogic (LTL) formulas containing the desired motion. We provide an automatedframework as the nominal solution to construct the hybrid controller that drives therobot such that its resulting trajectory satisfies the given task. Then we expand theproblem by considering a team of networked dynamical robots, where each robot hasa locally-specified individual task also as LTL formulas. In particular, we analyzefour different aspects as described below. When the workspace is only partially known to each robot, the nominal solutionmight be inadequate. Thus we first propose an algorithm for initial plan synthesis tohandle partially infeasible tasks that contain hard and soft constraints. We designan on-line scheme for each robot to verify and improve its local plan during runtime, utilizing its sensory measurements and communications with other robots. Itis ensured that the hard constraints for safety are always fulfilled while the softconstraints for performance are improved gradually. Secondly, we introduce a new approach to construct a full model of both robotmotion and actions. Based on this model, we can specify much broader robotic tasksand it is used to model inter-robot collaborative actions, which are essential for manymulti-robot applications to improve system capability, efficiency and robustness.Accordingly, we devise a distributed strategy where the robots coordinate theirmotion and action plans to fulfill the desired collaboration by their local tasks. Thirdly, continuous relative-motion constraints among the robots, such as collision avoidance and connectivity maintenance, are closely related to the stability,safety and integrity of multi-robot systems. We propose two different hybrid controlapproaches to guarantee the satisfaction of all local tasks and the relative-motionconstraints at all time: the first one is based on potential fields and nonlinear controltechnique; the second uses Embedded Graph Grammars (EGGs) as the main tool. At last, we take into account two common cooperative robotic tasks, namelyservice and formation tasks. These tasks are requested and exchanged among therobots during run time. The proposed hybrid control scheme ensures that the real-time plan execution incorporates not only local tasks of each robot but also thecontingent service and formation tasks it receives. Some of the theoretical results of the thesis have been implemented and demonstrated on various robotic platforms. / Denna avhandling fokuserar på distribuerad och hybridstyrning av multi-robot-system för komplexa, lokala och tidsberoende uppgifter. Dessa uppgifter specificerasav logiska formler rörande robotens rörelser och andra ageranden. Avhandlingenbehandlar ett tvärvetenskapligt område som integrerar reglering av nätverkaderobotsystem och planering baserad på formella metoder. Ett ramverk för hybridstyrning av flera dynamiska robotar med lokalt specificerade uppgifter presenteras.Fyra huvudscenarier betraktas: (1) robot-planering med motstridiga arbetsuppgifterinom ett delvis okänt arbetsområde; (2) beroende uppgifter för en grupp heterogenaoch samverkande robotar; (3) relativa rörelsebegränsningar hos varje robot; samt(4) robotar med uppgifter som begärs och bekräftas under körning. Numeriskasimuleringar och experiment visas för att validera de teoretiska resultaten. / <p>QC 20151204</p> / EU STREP RECONFIG: FP7-ICT-2011-9-600825 / Swedish Research Council (VR)

This is the final revision of my M.Sc. Thesis. / Runtime Verification is a technique to extract information from a running system in order to detect executions violating a given correctness specification. In this thesis, we study distributed synchronous/asynchronous runtime verification of systems. In our setting, there is a set of distributed monitors that have only partial views of a large system and are subject to failures. In this context, it is unavoidable that monitors may have different views of the underlying system, and therefore may have different valuations of the correctness property. In this thesis, we propose an automata-based synchronous monitoring algorithm that copes with f crash failures in a distrbuted setting. The algorithm solves the synchronous monitoring problem in f + 1 rounds of communication, and significantly reduces the message size overhead. We also propose an algorithm for distributed crash-resilient asynchronous monitoring that consistently monitors the system under inspection without any communication between monitors. Each local monitor emits a verdict set solely based on its own partial observation, and the intersection of the verdict sets will be the same as the verdict computed by a centralized monitor that has full view of the system. / Thesis / Master of Science (MSc)

Model checking is a fully-automatic formal verification method that has been extremely successful in validating and verifying safety-critical systems in the past three decades. In the past fifteen years, there has been a lot of work in extending many model checking algorithms over finite-state systems to finitely representable infinitestate systems. Unlike in the case of finite systems, decidability can easily become a problem in the case of infinite-state model checking. In this thesis, we present generic and specific techniques that can be used to derive decidability with near-optimal computational complexity for various model checking problems over infinite-state systems. Generic techniques and specific techniques primarily differ in the way in which a decidability result is derived. Generic techniques is a “top-down” approach wherein we start with a Turing-powerful formalismfor infinitestate systems (in the sense of being able to generate the computation graphs of Turing machines up to isomorphisms), and then impose semantic restrictions whereby the desired model checking problem becomes decidable. In other words, to show that a subclass of the infinite-state systems that is generated by this formalism is decidable with respect to the model checking problem under consideration, we will simply have to prove that this subclass satisfies the semantic restriction. On the other hand, specific techniques is a “bottom-up” approach in the sense that we restrict to a non-Turing powerful formalism of infinite-state systems at the outset. The main benefit of generic techniques is that they can be used as algorithmic metatheorems, i.e., they can give unified proofs of decidability of various model checking problems over infinite-state systems. Specific techniques are more flexible in the sense they can be used to derive decidability or optimal complexity when generic techniques fail. In the first part of the thesis, we adopt word/tree automatic transition systems as a generic formalism of infinite-state systems. Such formalisms can be used to generate many interesting classes of infinite-state systems that have been considered in the literature, e.g., the computation graphs of counter systems, Turing machines, pushdown systems, prefix-recognizable systems, regular ground-tree rewrite systems, PAprocesses, order-2 collapsible pushdown systems. Although the generality of these formalisms make most interesting model checking problems (even safety) undecidable, they are known to have nice closure and algorithmic properties. We use these nice properties to obtain several algorithmic metatheorems over word/tree automatic systems, e.g., for deriving decidability of various model checking problems including recurrent reachability, and Linear Temporal Logic (LTL) with complex fairness constraints. These algorithmic metatheorems can be used to uniformly prove decidability with optimal (or near-optimal) complexity of various model checking problems over many classes of infinite-state systems that have been considered in the literature. In fact, many of these decidability/complexity results were not previously known in the literature. In the second part of the thesis, we study various model checking problems over subclasses of counter systems that were already known to be decidable. In particular, we consider reversal-bounded counter systems (and their extensions with discrete clocks), one-counter processes, and networks of one-counter processes. We shall derive optimal complexity of various model checking problems including: model checking LTL, EF-logic, and first-order logic with reachability relations (and restrictions thereof). In most cases, we obtain a single/double exponential reduction in the previously known upper bounds on the complexity of the problems.

abstract: As robots become mechanically more capable, they are going to be more and more integrated into our daily lives. Over time, human’s expectation of what the robot capabilities are is getting higher. Therefore, it can be conjectured that often robots will not act as human commanders intended them to do. That is, the users of the robots may have a different point of view from the one the robots do. The first part of this dissertation covers methods that resolve some instances of this mismatch when the mission requirements are expressed in Linear Temporal Logic (LTL) for handling coverage, sequencing, conditions and avoidance. That is, the following general questions are addressed: * What cause of the given mission is unrealizable? * Is there any other feasible mission that is close to the given one? In order to answer these questions, the LTL Revision Problem is applied and it is formulated as a graph search problem. It is shown that in general the problem is NP-Complete. Hence, it is proved that the heuristic algorihtm has 2-approximation bound in some cases. This problem, then, is extended to two different versions: one is for the weighted transition system and another is for the specification under quantitative preference. Next, a follow up question is addressed: * How can an LTL specified mission be scaled up to multiple robots operating in confined environments? The Cooperative Multi-agent Planning Problem is addressed by borrowing a technique from cooperative pathfinding problems in discrete grid environments. Since centralized planning for multi-robot systems is computationally challenging and easily results in state space explosion, a distributed planning approach is provided through agent coupling and de-coupling. In addition, in order to make such robot missions work in the real world, robots should take actions in the continuous physical world. Hence, in the second part of this thesis, the resulting motion planning problems is addressed for non-holonomic robots. That is, it is devoted to autonomous vehicles’ motion planning in challenging environments such as rural, semi-structured roads. This planning problem is solved with an on-the-fly hierarchical approach, using a pre-computed lattice planner. It is also proved that the proposed algorithm guarantees resolution-completeness in such demanding environments. Finally, possible extensions are discussed. / Dissertation/Thesis / Doctoral Dissertation Computer Science 2019

Resilience is essential for long-term autonomous agents. Swarm behaviors seen in bees, ants, birds, fish, and others are interesting because they resiliently perform complex coordinated tasks like foraging, nest-selection, flocking and escaping predators without centralized control or coordination. Conventionally, mimicking swarm behaviors with robots requires researchers to study actual behaviors, derive mathematical models, and implement these models as state machines. Since the conventional approach is time-consuming and cumbersome, this dissertation uses a grammatical evolution algorithm with Behavior Trees (BTs) to evolve behaviors that are resilient to different perturbations for foraging and nest maintenance tasks. The modular, reactive, and readable properties of BTs make it an excellent controller for implementing swarm behaviors. Our method is based on the author's master's thesis work on a core algorithm called Grammatical Evolution algorithm for Evolution of Swarm bEhaviors using Behavior Trees (GEESE-BT). The GEESE-BT algorithm can be used to evolve swarm behaviors for interesting multiagent problems, but the solutions require ad hoc fitness functions tailored to the specific problems. This dissertation presents the BeTr-GEESE algorithm, which replaces ad hoc fitness functions with direct feedback from the BT modules. BeTr-GEESE learns more efficiently than GEESE-BT. The modular, subtask-specific programs produced by BeTr-GEESE can be exchanged through lateral transfer to perform missions that require sequential execution of subtasks. Lateral transfer produces resilient performance in divisible and additive group tasks like foraging and nest maintenance. However, the behaviors of successful groups must exhibit temporal locality, meaning that an agent must persist in behavior long enough to perform essential functions but also means that agents cannot persist too long or evolution is too slow. A biologically inspired enhancement of using multiple genes with BeTr-GEESE allowed a fixed population of heterogeneous agents to accomplish tasks with high resilience power and efficiency. The last part of the dissertation complements the empirical approach used in designing resilient swarms using grammatical evolution. Goal specification and verification are vital to designing resilient artificial agents. Finite trace Linear Temporal Logic ($LTL_f$) is a potent way of specifying goals, but synthesizing planners that guarantee the goals are satisfied can be computationally prohibitive. This dissertation shows that goals specified using a subset of finite trace Linear Temporal Logic ($LTL_f$) can be decomposed into an equivalent BT that leads to a relaxed behavior synthesis problem in which a wide range of planners can be used to generate effective behaviors that satisfy the goal specification.

Runtime Verification (RV) is usually considered as a lightweight automatic verification technique for the dynamic analysis of systems, where a monitor observes executions produced by a system and analyzes its executions against a formal specification. If the monitor were synthesized, in addition to the monitoring specification, also from extra assumptions on the system behavior (typically described by a model as transition systems), then it may output more precise verdicts or even be predictive, meanwhile it may no longer be lightweight, since monitoring under assumptions has the same computation complexity with model checking. When suitable assumptions come into play, the monitor may also support partial observability, where non-observable variables in the specification can be inferred from observables, either present or historical ones. Furthermore, the monitors are resettable, i.e. being able to evaluate the specification at non-initial time of the executions while keeping memories of the input history. This helps in breaking the monotonicity of monitors, which, after reaching conclusive verdicts, can still change its future outputs by resetting its reference time. The combination of the above three characteristics (assumptions, partial observability and resets) in the monitor synthesis is called the Assumption-Based Runtime Verification, or ABRV. In this thesis, we give the formalism of the ABRV approach and a group of monitoring algorithms based on specifications expressed in Linear Temporal Logic with both future and past operators, involving Boolean and possibly other types of variables. When all involved variables have finite domain, the monitors can be synthesized as finite-state machines implemented by Binary Decision Diagrams. With infinite-domain variables, the infinite-state monitors are based on satisfiability modulo theories, first-order quantifier elimination and various model checking techniques. In particular, Bounded Model Checking is modified to do its work incrementally for efficiently obtaining inconclusive verdicts, before IC3-based model checkers get involved. All the monitoring algorithms in this thesis are implemented in a tool called NuRV. NuRV support online and offline monitoring, and can also generate standalone monitor code in various programming languages. In particular, monitors can be synthesized as SMV models, whose behavior correctness and some other properties can be further verified by model checking.

The aim of this thesis is to explore the potential of resolution-based methods for linear temporal reasoning. On the abstract level, this means to develop new algorithms for automated reasoning about properties of systems which evolve in time. More concretely, we will: 1) show how to adapt the superposition framework to proving theorems in propositional Linear Temporal Logic (LTL), 2) use a connection between superposition and the CDCL calculus of modern SAT solvers to come up with an efficient LTL prover, 3) specialize the previous to reachability properties and discover a close connection to Property Directed Reachability (PDR), an algorithm recently developed for model checking of hardware circuits, 4) further improve PDR by providing a new technique for enhancing clause propagation phase of the algorithm, and 5) adapt PDR to automated planning by replacing the SAT solver inside with a planning-specific procedure. We implemented the proposed ideas and provide experimental results which demonstrate their practical potential on representative benchmark sets. Our system LS4 is shown to be the strongest LTL prover currently publicly available. The mentioned enhancement of PDR substantially improves the performance of our implementation of the algorithm for hardware model checking in the multi-property...

Cette thèse propose un nouveau type de systèmes de ré-écriture, appelé les systèmes de réécriture fonctionnels. Nous montrons que notre modèle a la puissance d'expression des systèmes de ré-écriture et qu'il est bien adapté à l'étude de propriétés de sûreté et de propriétés de logique temporelle de modèles.Nous avons mis en évidence une sous classe de systèmes fonctionnels, les élémentaires et les élémentaires à droite, préservant la puissance d'expression des systèmes fonctionnels et des techniques d'accélération des calculs aboutissant à un outil de vérification symbolique efficace.Dans la partie expérimentale, nous avons comparé notre outil, d'une part avec des outils de ré-écriture tels que Timbuk, Maude et TOM, d'autre part avec des outils de vérification tels que SPIN, NuSMV, SMART, HSDD. Nos benchmarks démontrent l'efficacité des systèmes fonctionnels élémentaires pour la vérification de modèles. / This PhD thesis proposes the theoretical foundations of a new formal tool for symbolic verification of finite models. Some approaches reduce the problem of system verification to the reachability problem in term rewriting systems (TRSs).In our approach, states are encoded by terms in a BDD-like manner and the transition relation is represented by a new rewriting relation so called Functional Term Rewriting Systems (FTRSs).First, we show that FTRSs are as expressive as TRSs. Then, we focus on a subclass of FTRSs, so called Elementary Functional Term Rewriting Systems (EFTRSs), and we show that EFTRSs preserve the FTRSs expressiveness. The main advantage of EFTRSs is that they are well adapted for acceleration techniques usually used in saturation algorithms on BDD-like data structures.Our experiments show that for well-known protocols (e.g. Tree Arbiter, Percolate, Round RobinMutex protocols,... ) our tool is not only better than other rewriting tools such as Timbuk, Maude or TOM, but also competitive with other model-checkers such as SPIN, NuSMV or SMART. Moreover, it can also be applied to model-checking invariant properties which are a particular subclass of linear temporal logic formula (LTL).

La vérification de logiciels est une activité dont l'importance est cruciale pour les logiciels embarqués critiques. Les différentes approches envisageables peuvent être classées en quatre catégories : les méthodes d'analyse statique non formelles, les méthodes d'analyse statique formelles, les méthodes d'analyse dynamique non formelles et les méthodes d'analyse dynamique formelles. L'objectif de cette thèse est de vérifier des propriétés temporelles dans un cadre industriel, par analyse dynamique formelle.La contribution comporte trois parties. Un langage adapté à l'expression des propriétés à vérifier, tirées du contexte industriel d'Airbus, a été dé ni. Il repose notamment sur la logique temporelle linéaire mais également sur un langage d'expressions régulières.La vérification d'une propriété temporelle s'effectue sur une trace d'exécution d'un logiciel, générée à partir d'un cas de test pré-existant. L'analyse statique est utilisée pour générer la trace en fonction des informations nécessaires à la vérification de la propriété temporelle formalisée.Cette approche de vérification propose une solution pragmatique au problème posé par le caractère ni des traces considérées. Des adaptations et des optimisations ont également été mises en œuvre pour améliorer l'efficacité de l'approche et faciliter son utilisation dans un contexte industriel. Deux prototypes ont été implémentés,des expérimentations ont été menées sur différents logiciels d'Airbus. / Software Verification is decisive for embedded software. The different verification approaches can be classified in four categories : non formal static analysis,formal static analysis, non formal dynamic analysis and formal dynamic analysis.The main goal of this thesis is to verify temporal properties on real industrial applications,with the help of formal dynamic analysis.There are three parts for this contribution. A language, which is well adapted to the properties we want to verify in the Airbus context was defined. This language is grounded on linear temporal logic and also on a regular expression language.Verification of a temporal property is done on an execution trace, generated from an existing test case. Generation also depends on required information to verify the formalized property. Static analysis is used to generate the trace depending on the formalized property.The thesis also proposes a pragmatic solution to the end of trace problem. In addition,specific adaptations and optimisations were defined to improve efficiency and user-friendliness and thus allow an industrial use of this approach. Two applications were implemented. Some experiments were led on different Airbus software.