Search by property

This page provides a simple browsing interface for finding entities described by a property and a named value. Other available search interfaces include the page property search, and the ask query builder.

List of results

• Synthesis of Control Protocols for Autonomous Systems  + (This article provides a review of control … This article provides a review of control protocol synthesis techniques that incorporate methodologies from formal methods and control theory to provide correctness guarantee for different types of autonomous systems, including those with discrete and continuous state space. The correctness of the system is defined with respect to a given specification expressed as a formula in linear temporal logic to precisely describe the desired properties of the system. The formalism presented in this article admits non-determinism, allowing uncertainties in the system to be captured. A particular emphasis is on alleviating some of the difficulties, e.g., heterogeneity in the underlying dynamics and computational complexity, that naturally arise in the construction of control protocols for autonomous systems. control protocols for autonomous systems.)
• Robotic Control and Nonholonomic Motion Planning  + (This dissertation addresses the problem of … This dissertation addresses the problem of control and kinematic</br>planning for constrained robot systems. An example of a system of</br>this type is a multifingered robot hand grasping an object. The</br>individual fingers act as robot manipulators and are constrained by</br>their contact with the object. If the contacts allow rolling between</br>the object and the fingertips, it is possible for the constraints to</br>be nonholonomic. That is, the constraints may not restrict the</br>reachable configurations of the system, but rather, constrain only the</br>allowable velocities of the system.</br><p></br>Using the multifingered hand as a motivating example, this</br>dissertation presents a detailed analysis of the kinematics, dynamics,</br>and control of robot systems with contact constraints. In particular,</br>it presents a unified derivation of the dynamics of robot manipulators</br>with Pfaffian velocity constraints, including the nonholonomic case.</br>This derivation allows control laws to be specified which are provably</br>stable for an entire class of systems, including unconstrained robots,</br>robot hands, and other systems of multiple robots performing a</br>coordinated task. A method for building complex controllers which</br>respects this class of constraints is also developed using a set of</br>simple primitives which allow hierarchical control structures to be</br>created in an organized fashion.</br><p></br>Finally, the nonholonomic motion planning problem is introduced and</br>discussed in detail. Using tools from differential geometric control</br>theory, it is possible to classify and analyze systems with</br>nonholonomic constraints. A brief review of the necessary tools along</br>with a review of the current literature is presented. A practical</br>method for steering nonholonomic systems using sinusoids is derived</br>and applied to several kinematic systems with contact constraints.to several kinematic systems with contact constraints.)
• Exponential Stabilization of Driftless Nonlinear Control Systems  + (This dissertation lays the foundation for … This dissertation lays the foundation for practical exponential</br>stabilization of driftless control systems. Driftless systems have</br>the form $$\dot x = X_1(x)u_1+\cdots +X_m(x)u_m, \quad x\in\real^n$$.</br>Such systems arise when modeling mechanical systems with nonholonomic</br>constraints. In engineering applications it is often required to</br>maintain the mechanical system around a desired configuration. This</br>task is treated as a stabilization problem where the desired</br>configuration is made an asymptotically stable equilibrium point. The</br>control design is carried out on an approximate system. The</br>approximation process yields a nilpotent set of input vector fields</br>which, in a special coordinate system, are homogeneous with respect to</br>a non-standard dilation. Even though the approximation can be given a</br>coordinate-free interpretation, the homogeneous structure is useful to</br>exploit: the feedbacks are required to be homogeneous functions and</br>thus preserve the homogeneous structure in the closed-loop system.</br>The stability achieved is called {\em $\rho$-exponential stability}.</br>The closed-loop system is stable and the equilibrium point is</br>exponentially attractive. This extended notion of exponential</br>stability is required since the feedback, and hence the closed-loop</br>system, is not Lipschitz. However, it is shown that the convergence</br>rate of a Lipschitz closed-loop driftless system cannot be bounded by</br>an exponential envelope.</br><p></br>The synthesis methods generate feedbacks which are smooth on</br>\rminus. The solutions of the closed-loop system are proven to be</br>unique in this case. In addition, the control inputs for many</br>driftless systems are velocities. For this class of systems it is</br>more appropriate for the control law to specify actuator forces</br>instead of velocities. We have extended the kinematic velocity</br>controllers to controllers which command forces and still</br>$\rho$-exponentially stabilize the system.</br><p></br>Perhaps the ultimate justification of the methods proposed in this</br>thesis are the experimental results. The experiments demonstrate the</br>superior convergence performance of the $\rho$-exponential stabilizers</br>versus traditional smooth feedbacks. The experiments also highlight</br>the importance of transformation conditioning in the feedbacks. Other</br>design issues, such as scaling the measured states to eliminate</br>hunting, are discussed. The methods in this thesis bring the</br>practical control of strongly nonlinear systems one step closer.</br><p>rol of strongly nonlinear systems one step closer. <p>)
• Model-Based Estimation of Off-Highway Road Geometry using Single-Axis LADAR and Inertial Sensing  + (This paper applies some previously studied … This paper applies some previously studied extended</br>Kalman filter techniques for planar road geometry</br>estimation to the domain of autonomous navigation of offhighway</br>vehicles. In this work, a clothoid model of the road</br>geometry is constructed and estimated recursively based on</br>road features extracted from single-axis LADAR range measurements.</br>We present a method for feature extraction of the</br>road centerline in the image plane, and describe its application</br>to recursive estimation of the road geometry. We analyze</br>the performance of our method against simulated motion of</br>varied road geometries and against recorded data from previous</br>autonomous navigation runs. Our method accomodates full 6</br>DOF motion of the vehicle as it navigates, constructs consistent</br>estimates of the road geometry with respect to a fixed global</br>reference frame, and requires an estimate of the sensor pose</br>for each range measurement.he sensor pose for each range measurement.)
• Receding Horizon Temporal Logic Planning for Dynamical Systems  + (This paper bridges the advances in compute … This paper bridges the advances in computer science and control to allow automatic synthesis of complex dynamical systems which are guaranteed, by construction, to satisfy the desired properties even in the presence of adversary. The desired properties are expressed in the language of temporal logic. With its expressive power, a wider class of properties than safety and stability can be specified. The resulting system consists of a discrete planner which plans, in the abstracted discrete domain, a set of transitions of the system to ensure the correct behaviors and a continuous controller which continuously implements the plan. For a system with certain structure, we present an approach, based on a receding horizon scheme, to overcome computational difficulties in the synthesis of a discrete planner and allow more complex problems to be solved. allow more complex problems to be solved.)
• Exponential Stabilization of Driftless Nonlinear Control Systems via Time-varying, Homogeneous Feedback  + (This paper brings together results from a … This paper brings together results from a number of different areas in</br>control theory to provide an algorithm for the synthesis of locally</br>exponentially stabilizing control laws for a large class of driftless</br>nonlinear control systems. The exponential stabilization relies on the</br>use of feedbacks which render the closed loop vector field homogeneous</br>with respect to a dilation. These feedbacks are generated from a</br>modification of Pomet's algorithm for smooth feedbacks. Converse</br>Liapunov theorems for time-periodic homogeneous vector fields</br>guarantee that local exponential stability is maintained in the</br>presence of higher order (with respect to the dilation) perturbing</br>terms.</br><p>ect to the dilation) perturbing terms. <p>)
• Real-valued average consensus over noisy quantized channels  + (This paper concerns the average consensus … This paper concerns the average consensus problem with the constraint of quantized communication between nodes. A broad class of algorithms is analyzed, in which the transmission strategy, which decides what value to communicate to the neighbors, can include various kinds of rounding, probabilistic quantization, and bounded noise. The arbitrariness of the transmission strategy is compensated by a feedback mechanism which can be interpreted as a self-inhibitory action. The result is that the average of the nodes state is not conserved across iterations, and the nodes do not converge to a consensus; however, we show that both errors can be made as small as desired. Bounds on these quantities involve the spectral properties of the graph and can be proved by employing elementary techniques of LTI systems analysis.entary techniques of LTI systems analysis.)
• Decentralised minimal-time dynamic consensus  + (This paper considers a group of agents tha … This paper considers a group of agents that aim to reach an agreement on individually measured time-varying signals by local communication. In contrast to static network averaging problem, the consensus we mean in this paper is reached in a dynamic sense. A discrete-time dynamic average consensus protocol can be designed to allow all the agents tracking the average of their reference inputs asymptotically. We propose a minimal-time dynamic consensus algorithm, which only utilises minimal number of local observations of randomly picked node in a network to compute the final consensus signal. Our results illustrate that with memory and computational ability, the running time of distributed averaging algorithms can be indeed improved dramatically using local information as suggested by Olshevsky and Tsitsiklis. as suggested by Olshevsky and Tsitsiklis.)
• Stability and Performance of Non-Homogeneous Multi-Agent Systems on a Graph  + (This paper considers distributed control o … This paper considers distributed control of interconnected multi-agent systems. The dynamics of the individual agents are not required to be homogeneous and the interaction topology is described by an arbitrary directed graph. We derive the sensitivity transfer functions between every pair of agents and we analyze stability and performance of non-homogeneous systems, showing that the low frequency behavior is influenced not only by topology, but also by static gain and poles of the agents.so by static gain and poles of the agents.)
• On Decentralized Classification using a Network of Mobile Sensors  + (This paper considers how a team of mobile … This paper considers how a team of mobile sensors</br>should cooperatively move so as to optimally categorize a</br>single moving target from their noisy sensor readings. The</br>cooperative control procedure is based on the development of</br>a cost function that quantifies the teamâs classification error.</br>The robotsâ motions are then chosen to minimize this function.</br>We particularly investigate the case where the sensor noise and</br>class distributions are Gaussian. In this case, we can derive a</br>duality principle which states that optimal classification will be</br>realized when the covariance of the target estimate is minimized.</br>That is, in this case, optimal estimation leads naturally to</br>optimal classification. We extend previous work to develop a</br>distributed discrete-gradient search algorithm that guides the</br>teamâs location motions for purposes of optimal estimation and</br>classification. The concepts developed are validated through</br>numerical studies.d are validated through numerical studies.)
• Nonlinear Control Methods for Planar Carangiform Robot Fish  + (This paper considers the design of motion … This paper considers the design of motion control algorithms for robot fish. We present modeling, control design, and experimental trajectory tracking results for a planar robotic fish that is propelled using the carangiform style of locomotion. Our experimental apparatus consists of a freely translating and rotating flat plate and a two-link actuated tail. We develop a model for the fish's propulsion that is based on quasi-steady fluid flow. Using this model, we predict system response to sinusoidal motions of the tail joints and</br>compare these predictions to the experimental results. We then propose gaits for forward and turning trajectories and analyze system response under such control strategies. Our models and predictions are verified by experiment.nd predictions are verified by experiment.)
• A Testbed for Nonlinear Flight Control Techniques: The Caltech Ducted Fan  + (This paper considers the fundamental desig … This paper considers the fundamental design and modeling of the Caltech ducted fan.</br>The Caltech ducted fan is a scaled model of the longitudinal axis of a flight vehicle. The</br>purpose of the ducted fan is the research and development of new nonlinear flight guidance</br>and control techniques for Uninhabited Combat Aerial Vehicles. It is shown that critical</br>design relations must be satisfied in order that the ducted fan's longitudinal dynamics</br>behave similar to those of an flight vehicle. Preliminary flight test results illustrate</br>the flying qualities of the ducted fan.te the flying qualities of the ducted fan.)
• Estimation over Communication Networks: Performance Bounds and Achievability Results  + (This paper considers the problem of estima … This paper considers the problem of estimation over</br>communication networks. Suppose a sensor is taking</br>measurements of a dynamic </br>process. However the process needs to be estimated at a</br>remote</br>location connected to the sensor through a network of</br>communication links that drop</br>packets stochastically. We provide a framework for</br>computing the</br>optimal performance in the sense of expected error</br>covariance. Using</br>this framework we characterize the dependency of the</br>performance on</br>the topology of the network and the packet dropping</br>process. For</br>independent and memoryless packet dropping processes we</br>find the</br>steady-state error for some classes of networks and obtain</br>lower and</br>upper bounds for the performance of a general network. We</br>also</br>illustrate how this framework can be used in the synthesis</br>of</br>networks for the purpose of estimation. Finally we find a</br>necessary</br>and sufficient condition for the stability of the estimate</br>error</br>covariance for general networks with spatially correlated</br>and Markov</br>type dropping process. This interesting condition has a</br>max-cut</br>interpretation.ng condition has a max-cut interpretation.)
• A Motion Planner for Nonholonomic Robots  + (This paper considers the problem of motion … This paper considers the problem of motion planning for a car-like robot (i.e., a</br>mobile robot with a nonholonomic constraint whose turning radius is lower-bounded). We</br>present a fast and exact planner for our mobile robot model, based upon recursive</br>subdivision of a collision-free path generated by a lower-level geometric planner that</br>ignores the motion constraints. The resultant trajectory is optimized to give a path that</br>is of near-minimal length in its homotopy class. Our claims of high speed are supported by</br>experimental results for implementations that assume a robot moving amid polygonal</br>obstacles. The completeness and the complexity of the algorithm are proven using an</br>appropriate metric in the configuration space R2 x S1 of the robot. This metric is defined</br>by using the length of the shortest paths in the absence of obstacles as the distance</br>between two configurations. We prove that the new induced topology and the classical one</br>are the same. Although we concentration upon the car-like robot, the generalization of</br>these techniques leads to new theoretical issues involving sub-Riemannian geometry and to</br>practical results for nonholonomic motion planning. results for nonholonomic motion planning.)
• Real Time Trajectory Generation for Differentially Flat Systems  + (This paper considers the problem of real t … This paper considers the problem of real time trajectory generation and tracking for</br>nonlinear control systems. We employ a two degree of freedom approach that separates the</br>nonlinear tracking problem into real time trajectory generation followed by local</br>(gain-scheduled) stabilization. The central problem which we consider is how to generate,</br>possibly with some delay, a feasible state space and input trajectory in real time from an</br>output trajectory that is given online. We propose two algorithms that solve the real time</br>trajectory generation problem for differentially flat systems with (possibly non-minimum</br>phase) zero dynamics. One is based on receding horizon point to point steering, the other</br>allows additional minimization of a cost function. Both algorithms explicitly address the</br>tradeoff between stability and performance and we prove convergence of the algorithms for</br>a reasonable class of output trajectories. To illustrate the application of these</br>techniques to physical systems, we present experimental results using a vectored thrust</br>flight control experiment built at Caltech. A brief introduction to differentially flat</br>systems and its relationship with feedback linearization is also included.h feedback linearization is also included.)
• Synthesis of Correct-by-construction Control Protocols for Hybrid Systems Using Partial State Information  + (This paper considers the problem of synthe … This paper considers the problem of synthesizing output-feedback control laws for a class of discrete-time hybrid systems in order for the trajectories of the system to satisfy certain high-level specifications expressed in linear temporal logic. By leveraging ideas from robust interpretation of temporal logic formulas and bounded-error estimation, we identify a subclass of systems for which it is possible to reduce the problem to a state-feedback form. In particular, we use locally superstable hybrid observers to resolve the partial information at the continuous level. This allows us to use recent results in temporal logic planning to synthesize the desired controllers based on two-player perfect- information games. The overall control architecture consists of a hybrid observer, a high-level switching protocol and a low-level continuous controller. We demonstrate the proposed framework in a case study on designing control protocols for an aircraft air management system.ols for an aircraft air management system.)
• Backtracking temporal logic synthesis for uncertain environments  + (This paper considers the problem of synthe … This paper considers the problem of synthesizing correct-by-construction robotic controllers in environments with uncertain but fixed structure. âEnvironmentâ has two notions in this work: a map or âworldâ in which some controlled agent must operate and navigate (i.e. evolve in a configuration space with obstacles); and an adversarial player that selects con- tinuous and discrete variables to try to make the agent fail (as in a game). Both the robot and the environment are subjected to behavioral specifications expressed as an assume-guarantee linear temporal logic (LTL) formula. We then consider how to efficiently modify the synthesized controller when the robot encounters unexpected changes in its environment. The crucial insight is that a portion of this problem takes place in a metric space, which provides a notion of nearness. Thus if a nominal plan fails, we need not resynthesize it entirely, but instead can âpatchâ it locally. We present an algorithm for doing this, prove soundness, and demonstrate it on an example gridworld.nd demonstrate it on an example gridworld.)
• On Quantized Consensus by Means of Gossip Algorithm -- Part II: Convergence Time  + (This paper deals with the distributed aver … This paper deals with the distributed averaging problem over a connected network of agents, subject to a quantization constraint. It is assumed that at each time update, only a pair of agents can update their own numbers in terms of the quantized data being exchanged. The agents are also required to communicate with one another in a stochastic fashion. In the first part of the paper, it was shown that the quantized consensus is reached by means of a stochastic gossip algorithm proposed in a recent paper, for any arbitrary quantization. The current part of the paper considers the expected value of the time at which the quantized consensus is reached. This quantity (corresponding to the worst case) is upper and lower bounded in terms of the topology of the graph, for uniform quantization. In particular, it is shown that these bounds are related to the principal minors of the weighted Laplacian matrix. A convex optimization is also proposed to determine the set of probabilities (used to pick a pair of agents) which leads to the fast convergence of the gossip algorithm. fast convergence of the gossip algorithm.)
• Tuning a synthetic in vitro oscillator using control-theoretic tools  + (This paper demonstrates the effectiveness … This paper demonstrates the effectiveness of sim- ple control-theoretic tools in generating simulation-guided ex- periments on a synthetic in vitro oscillator. A theoretical analysis of the behavior of such system is motivated by high cost, time consuming experiments, together with the excessive number of tuning parameters. A simplified model of the synthetic oscillator is chosen to capture only its essential features. The model is analyzed using the small gain theorem and the theory of describing functions. Such analysis reveals what are the parameters that primarily determine when the system can admit stable oscillations. Experimental verification of the theoretical and numerical findings is carried out and confirms the predicted results regarding the role of production and degradation rates. role of production and degradation rates.)
• TuLiP: A Software Toolbox for Receding Horizon Temporal Logic Planning  + (This paper describes TuLiP, a Python-based … This paper describes TuLiP, a Python-based software tool- box for the synthesis of embedded control software that is provably correct with respect to an expressive subset of lin- ear temporal logic (LTL) specifications. TuLiP combines routines for (1) finite state abstraction of control systems, (2) digital design synthesis from LTL specifications, and (3) receding horizon planning. The underlying digital de- sign synthesis routine treats the environment as adversary; hence, the resulting controller is guaranteed to be correct for any admissible environment profile. TuLiP applies the re- ceding horizon framework, allowing the synthesis problem to be broken into a set of smaller problems, and consequently alleviating the computational complexity of the synthesis procedure, while preserving the correctness guarantee.hile preserving the correctness guarantee.)
• Design of insulating devices for in vitro synthetic circuits  + (This paper describes a synthetic in vitro … This paper describes a synthetic in vitro genetic circuit programmed to work as an insulating device. This circuit is composed of nucleic acids, which can be designed to interact according to user defined rules, and of few proteins that perform catalytic functions. A model of the circuit is derived from first principle biochemical laws. This model is shown to exhibit time-scale separation that makes its output insensitive to downstream time varying loads. Simulation results show the circuit effectiveness and represent the starting point for future experimental testing of the device.future experimental testing of the device.)
• Reduction of Bleed Valve Bandwidth Requirements for Control of Rotating Stall using Continuous Air Injection  + (This paper describes a technique for reduc … This paper describes a technique for reducing the actuator rate</br>requirement in active control of rotating stall and surge on</br>compressors. Actuation of the compressor characteristic, via the use</br>of continuous air injection at the rotor face in a single-stage, low</br>speed axial compressor, has been experimentally verified to provide a</br>method of reducing the actuator rate requirement of bleed valve</br>control of rotating stall. With a compressor rotor frequency of 100</br>Hz, active control of stall with a high speed bleed valve is achieved</br>only when the compressor characteristic is actuated. Furthermore, the</br>experiments show that the bleed valve rate requirement is reduced from</br>approximately 145 Hz to below 10 Hz when the amount of compressor</br>characteristic actuation is increased. Theoretical tools based on a</br>low order model (1-3 states) and simulations based on a reduced order</br>distributed model (37 states) have been developed to estimate the gain</br>and rate requirements of the bleed controller. All of the analytical</br>formulas and simulations share the same qualitative trends with</br>respect to the second and third derivatives of the compressor</br>characteristic function evaluated at the peak, and the experiments.</br>The agreement implies that bleed valve control of rotating stall</br>depends crucially on the rate limit of the bleed valve which in turn</br>depends on both the stable and the unstable part of the compressor</br>characteristic. Actuation of the compressor characteristic is</br>concluded to be a valuable tool in circumventing the rate limitation</br>of bleed valves for control of stall. By combining the compressor</br>characteristic identification tools and the analytic relations,</br>insights for designing a compressor-bleed pair are provided.ning a compressor-bleed pair are provided.)
• Approximate Trajectory Generation for Differentially Flat Systems with Zero Dynamics  + (This paper describes algorithms to generat … This paper describes algorithms to generate trajectories </br>for differentially flat systems with zero dynamics. </br>Zero dynamics in flat systems occur when the flat outputs </br>are not the tracking outputs. This means that the output </br>trajectories can be fully parametrized by the flat outputs, </br>but that there is some additional freedom left. </br>This freedom can be exploited to minimize a cost criterion. </br>We parametrize the differentially </br>flat outputs by basis functions, and solve for the parameters </br>so as to track a prescribed trajectory approximately </br>while minimizing a cost function. </br>We focus on implementation issues and point out the computational </br>cost involved in the various problems.al cost involved in the various problems.)
• Extending Exponential Stabilizers for Nonholonomic Systems from Kinematic Controllers to Dynamic Controllers  + (This paper describes an approach for exten … This paper describes an approach for extending (time-varying)</br>exponential stabilizers for nonholonomic systems from controllers which</br>command input velocity to controllers which command input torques. Due</br>to the nondifferentiable nature of exponential stablizers, additional</br>structure is required in order to ensure that the extended controllers</br>generate continuous control actions. In this paper we show how to</br>extend homogeneous controllers which use a nonstandard dilation adapted</br>to the problem.nstandard dilation adapted to the problem.)
• Differential Flatness of Mechanical Control Systems: A Catalog of Prototype Systems  + (This paper describes the application of di … This paper describes the application of differential flatness</br>techniques from nonlinear control theory to mechanical (Lagrangian)</br>systems. Systems which are differentially flat have several useful</br>properties which can be exploited to generate effective control</br>strategies for nonlinear systems. For the special case of mechanical</br>control systems, much more geometric information is present and the</br>purpose of this paper is to explore the implications and features of</br>that class of systems. We concentrate on several worked examples</br>which illustrate the general theory and present a detailed catalog of</br>known examples of differentially flat mechanical systems.of differentially flat mechanical systems.)

• Design and Implementation of a Biomolecular Circuit for Tracking Protein Concentration  + (This paper describes the design, modeling, … This paper describes the design, modeling, synthesis and preliminary validation of a protein concentration regulator circuit. The circuit is designed to maintain the level of a target protein to a reference level, specified by the amount of another protein. This is implemented using a single negative feedback loop that inhibited the production of the target protein once its concentration was equal to the reference amount. A mathematical model consisting of a set of ODEs is derived from mass action laws and Hill function approximations of protein production. Steady-state analysis of the model is used to predict parameter sensitivity and experimental behavior. We implemented this circuit in ''E. coli'' using scaffold-based sequestration and transcriptional activation. Preliminary experimental results show the system matching predictions from our model and performing the expected task.ur model and performing the expected task.)
• Interfacing TuLiP with the JPL Statechart Autocoder: Initial progress toward synthesis of flight software from formal specifications  + (This paper describes the implementation of … This paper describes the implementation of an in- terface connecting the two tools : the JPL SCA (Statechart Autocoder) and TuLiP (Temporal Logic Planning Toolbox) to enable the automatic synthesis of low level implementation code directly from formal specifications. With system dynamics, bounds on uncertainty and formal specifications as inputs, TuLiP synthesizes Mealy machines that are correct-by-construction. An interface is built that automatically translates these Mealy machines into UML statecharts. The SCA accepts the UML statecharts (as XML files) to synthesize flight-certified2 implementation code. The functionality of the interface is demonstrated through three example systems of varying com- plexity a) a simple thermostat b) a simple speed controller for an autonomous vehicle and c) a more complex speed controller for an autonomous vehicle with a map-element. In the thermostat controller, there is a specification regarding the desired temperature range that has to be met despite disturbance from the environment. Similarly, in the speed-controllers there are specifications about safe driving speeds depending on sensor health (sensors fail unpredictably) and the map-location. The significance of these demonstrations is the potential circumventing of some of the manual design of statecharts for flight software/controllers. As a result, we expect that less testing and validation will be necessary. In applications where the products of synthesis are used alongside manually designed components, extensive testing or new certificates of correctness of the composition may still be required. of the composition may still be required.)
• Alice: An information-rich autonomous vehicle for high-speed desert navigation  + (This paper describes the implementation an … This paper describes the implementation and testing of Alice, the California Institute of Technology’s entry in the 2005 DARPA Grand Challenge. Alice utilizes a highly networked control system architecture to provide high performance, autonomous driving in unknown en- vironments. Innovations include a vehicle architecture designed for efficient testing in harsh environments, a highly sensory-driven approach to fuse sensor data into speed maps used by real-time trajectory optimization algorithms, health and contingency management algorithms to manage failures at the component and system level, and a software logging and display envi- ronment that enables rapid assessment of performance during testing. The system successfully completed several runs in the National Qualifying Event, but encountered a combination of sens- ing and control issues in the Grand Challenge Event that led to a critical failure after traversing approximately 8 miles.re after traversing approximately 8 miles.)
• A Domain-Specific Language for Reactive Control Protocols for Aircraft Electric Power Systems  + (This paper describes the use of a domain-s … This paper describes the use of a domain-specific language, and an accompanying software tool, in constructing correct- by-construction control protocols for aircraft electric power systems. Given a base topology, the language consists of a set of primitives for standard specifications. The accompanying tool converts these primitives into formal specifica- tions, which are used to synthesize control protocols. We can then use TuLiP, a Python-based software toolbox, to synthesize centralized and distributed controllers. For sys- tems with no time involved in the specifications, this tool also provides an option to output specifications into a SAT-solver compatible format, thus reducing the synthesis problem to a satisfiability problem. We provide the results of our synthesis procedure on a range of topologies.thesis procedure on a range of topologies.)
• Controlling biological networks by time-delayed signals  + (This paper describes the use of time-delay … This paper describes the use of time-delayed feedback to regulate the behavior of biological networks. The general ideas are demonstrated on specific transcriptional regulatory and neural networks. It is shown that robust yet tunable controllers can be constructed that provide the biological systems with model-engineered inputs. The results indicate that time delay modulation may serve as an efficient bio-compatible control tool. an efficient bio-compatible control tool.)
• Receding Horizon Control of a Vectored Thrust Flight Experiment  + (This paper details the application of a co … This paper details the application of a constrained receding horizon control techniques to</br>stabilize an indoor vectored-thrust flight experiment, known as the Caltech Ducted Fan, subjected to step commands. The result of the experiment sheds light on the theoretical stability</br>of the receding horizon control strategy as well as validates some efficient computation techniques for solving nonlinear optimal control problems with constraints. The receding horizon</br>control problem is formulated as a constrained optimal control problem, which was solved with</br>an e±cient computational method developed and implemented by Milam et al., that combines</br>nonlinear control theory, B-spline basis functions, and nonlinear programming. Characteristic</br>issues including non-zero computational times, convergence property, choice of horizon length</br>and terminal cost are discussed. The study confirms the applicability of real-time receding</br>horizon control for constrained systems with fast dynamics.or constrained systems with fast dynamics.)
• Real-Time Trajectory Generation for the Cooperative Path Planning of Multi-Vehicle Systems  + (This paper discusses a Cooperative Path Pl … This paper discusses a Cooperative Path Planning (CPP) design</br>methodology for multi-vehicle systems and a Nonlinear</br>Trajectory Generation (NTG) algorithm. Three scenarios</br>of multi-vehicle tasking are proposed at the CPP framework.</br>The NTG algorithm is, then, used to generate realtime</br>trajectory for desired vehicle activities. Given system</br>dynamics and constraints, the NTG algorithm first finds trajectory</br>curves in a lower dimensional space and, then, parameterizes</br>the curves by the B-spline basis. The coefficients of the B-splines are further solved by the sequential</br>quadratic programming to satisfy the optimization objectives</br>and constraints. The NTG algorithm has been implemented</br>to generate real-time trajectories for a group of</br>cooperative vehicles in the presence of changing missions</br>and constraints.ence of changing missions and constraints.)
• Cooperative Task Planning of Multi-Robot Systems with Temporal Constraints  + (This paper discusses a design methodology … This paper discusses a design methodology of cooperative</br>trajectory generation for multi-robot systems.</br>The trajectory of achieving cooperative tasks, i.e., with</br>temporal constraints, is constructed by a nonlinear trajectory</br>generation (NTG) algorithm. Three scenarios</br>of multi-robot tasking are proposed at the cooperative</br>task planning framework. The NTG algorithm is, then,</br>used to generate real-time trajectory for desired robot</br>activities. Given robot dynamics and constraints, the</br>NTG algorithm first finds trajectory curves in a lower</br>dimensional space and, then, parameterizes the curves</br>by a set of B-spline representations. The coe�cients</br>of the B-splines are further solved by the sequential</br>quadratic programming to satisfy the optimization objectives</br>and constraints. The NTG algorithm has been</br>implemented to generate real-time trajectories for a</br>group of cooperative robots in the presence of spatial</br>and temporal constraints. Finally, an illustrated example</br>of cooperative task planning with temporal constraints</br>is presented.ng with temporal constraints is presented.)
• Low-Observable Nonlinear Trajectory Generation for Unmanned Air Vehicles  + (This paper explores low observability flig … This paper explores low observability flight path planning of unmanned air vehicles (UAVs)</br>in the presence of radar detection systems. The probability of detection model of an aircraft</br>near an enemy radar depends on aircraft attitude, range, and configuration. A detection model</br>is coupled with a simplified aircraft dynamics model. The Nonlinear Trajectory Generation</br>(NTG) software package developed at Caltech is used. The NTG algorithm is a gradient descent</br>optimization method that combines three technologies: Bsplines, output space collocation and</br>nonlinear optimization tools. Implementations are formulated with temporal constraints that</br>allow periods of high observability interspersed with periods of low observability. Illustrative</br>examples of optimized routes for low observability are presented.outes for low observability are presented.)
• A Framework for Low--Observable Tra jectory Generation in the Presence of Multiple Radars  + (This paper explores the problem of finding … This paper explores the problem of finding a real--time optimal tra jectory for unmanned air vehicles (UAV) in order to minimize their probability of detection by opponent multiple radar detection systems. The problem is handled using the Nonlinear Tra jectory Generation (NTG) method developed by Milam et al. The paper presents a formulation of the trajectory generation task as an optimal control problem, where temporal constraints allow periods of high observability interspersed with periods of low observability. This feature can be used strategically to aid in avoiding detection by an opponent radar. The guidance is provided in the form of sampled tabular data. It is then shown that the success of NTG on the proposed low--observable tra jectory generation problem depends upon an accurate parameterization of the guidance data. In particular, such an approximator is desired to have a compact architecture, a minimum number of design parameters, and a smooth continuously--differentiable input-output mapping. Artificial Neural Networks (ANNs) as universal approximators are known to possess these features, and thus are considered here as appropriate candidates for this task. Comparison of ANNs against B-spline approximators is provided, as well. Numerical simulations on multiple radar scenarios illustrate UAV </br>trajectories optimized for both detectability and time.optimized for both detectability and time.)
• Stability Analysis of Interconnected Nonlinear Systems Under Matrix Feedback  + (This paper explores the stability analysis … This paper explores the stability analysis problem for nonlinear systems which have general</br>linear feedback interconnections. Systems are often modeled in this manner in the study of</br>decentralized control because many communication topologies can be modeled and analyzed</br>using connections to graph theory. We present necessary conditions for stability of a classification</br>of interconnected systems, and we give some examples to provide insight into this problem.</br>These conditions are related to positive definiteness of matrices associated with the feedback</br>interconnection, and specialize to the common case where the Laplacian matrix of a graph</br>represents the communication topology of the system. the communication topology of the system.)
• Limits on the network sensitivity function for multi-agent systems on a graph  + (This paper explores the tradeoffs and limi … This paper explores the tradeoffs and limits of performance in feedback control of interconnected multi-agent systems, focused on the network sensitivity functions. We consider the interaction topology described by a directed graph and we prove that the sensitivity transfer functions between every pair of agents, arbitrarily connected, can be derived using a version of the Mason's Direct Rule. Explicit forms for special types of graphs are presented. An analysis of the role of cycles points out that these structures influence and limit considerably the behavior of the system. The more the cycles are equally distributed among the formation, the better performance the system can achieve, but they are always worse than the single agent case. We also prove the networked version of Bode's integral formula, showing that it still holds for multi-agent systems.at it still holds for multi-agent systems.)
• Feedback architectures to regulate flux of components in artificial gene networks  + (This paper focuses on RNA flux regulation … This paper focuses on RNA flux regulation for in vitro synthetic gene networks and considers architectures that can be scaled to an arbitrary number of species. Feedback loops are designed based on negative autoâregulation (which can minimize the potentially harmful amount of molecules not used to form useful products) and crossâactivation (which can maximize the overall output flux): transcription rate matching can be achieved through proper feedback constants; negative feedback is faster and maintains stability. A possible experimen- tal implementation of a three and four genes negative feedback architecture is also numerically studied. architecture is also numerically studied.)
• Control on the Sphere and Reduced Attitude Stabilization  + (This paper focuses on a new geometric appr … This paper focuses on a new geometric approach to fully actuated control</br>systems on the Riemannian manifold S^2. Our control laws exploit the basic</br>and intuitive notions of geodesic direction and of distance between points,</br>and generalize the classical proportional plus derivative feedback (PD)</br>without the need of arbitrary local coordinate charts. Even for the</br>stability analysis, the appropriate Lyapunov function relies upon the notion</br>of distance and its properties. This methodology then applies to spin-axis</br>stabilization of a spacecraft actuated by only two control torques:</br>discarding the rotation about the unactuated axis, a reduced system is</br>considered, whose state is in fact defined on the sphere. For this reduced</br>attitude stabilization problem our approach allows us not only to deal</br>optimally with the inevitable singularity, but also to achieve simplicity,</br>versatility and (coordinate independent) adaptive capabilities.dinate independent) adaptive capabilities.)
• Real-Time Control Experiments for Instruction and Research at Caltech  + (This paper gives a brief introduction to s … This paper gives a brief introduction to some of the experimental</br>facilities at Caltech being used for instruction and research. Two</br>experiments are described: a flight control experiment using a vector</br>thrust engine, and a low-speed, axial flow compressor rig. A common</br>hardware and software infrastructure is used to control these</br>experiments, allowing easy transition from instructional to research</br>use.sition from instructional to research use.)
• Trajectory Generation for Mechanical Systems with Application to Robotic Locomotion  + (This paper gives a survey of current and e … This paper gives a survey of current and emerging techniques for motion control of</br>nonlinear mechanical systems, motivated by applications in robotic locomotion. For this</br>class of systems, internal changes of shape and/or application of body fixed forces are</br>the mechanism by which the robot mov es in its environment and the geometric mechanical</br>properties of the system are crucial in understanding how specific gaits for generating</br>motion can be obtained.its for generating motion can be obtained.)
• Geometric Approaches to Control in the Presence of Magnitude and Rate Saturations  + (This paper gives a survey of some recent r … This paper gives a survey of some recent results on control of systems with magnitude</br>and rate limits, motivated by problems in real-time trajectory generation and tracking for</br>unmanned aerial vehicles. Two problems are considered: stabilization using ``nonlinear</br>wrappers'' to rescale a given control law and real-time trajectory generation using</br>differential flatness. For both problems, simplified versions of the general problem are</br>studied using tools from differential geometry and nonlinear control to give insights into</br>the limitations imposed by magnitude and rate limits and provide insights into</br>constructive solutions to the trajectory generation and tracking problems.ajectory generation and tracking problems.)
• Uncertainty-based Sensor Fusion of Range Data for Real-time Digital Elevation Mapping (RTDEM)  + (This paper introduces a new computationall … This paper introduces a new computationally</br>inexpensive approach to perception and modeling of the</br>environment that allows fusion of sensory range data of</br>various types and fidelities while explicitly taking into account</br>a complete description of uncertainty of the range measurements.</br>This approach makes use of known sensor uncertainty</br>models to create a single 2.5D digital elevation map whose</br>accuracy is robust to sensor noise and spurious data. This</br>approach is particularly suitable for real-time application in</br>high speed and highly unstructured outdoor environments</br>for which reasonably accurate and timely vehicle state estimates</br>are available. Experimental results are presented in</br>which LADAR range measurements and state estimates are</br>combined according to this approach. We provide qualitative</br>comparison to other classes of environment modeling. to other classes of environment modeling.)
• Outer flatness: trajectory generation for a model helicopter  + (This paper introduces the concept of outer … This paper introduces the concept of outer flatness, </br>a derivative of differential flatness. </br>Outer flatness describes a </br>system that can be split </br>in 2 subsytems, a non-flat inner system and a flat outer system. </br>The outputs of the outer system are the tracking outputs of interest. </br>The inputs of the outer system are the outputs of the inner system, </br>and not subject to our direct control. </br>The inputs of the inner system are the real actuator inputs. </br>This system structure is also present in backstepping and dynamic</br>inversion. </br>We present two theorems on exponential and bounded tracking for </br>outer flat systems, based on Lyapunoff arguments. </br>We validate the approach with simulations and experiments </br>on a model helicopter. </br><p>d experiments on a model helicopter. <p>)
• Stabilization of Feedback Systems via Distribution of Delays  + (This paper investigates the results of dis … This paper investigates the results of distributing the delay of a single feedback system. To distribute the delayed feedback, the single delay is replaced by the sum of two distinct delays with the same effective delay. The statistical properties of the new distribution function in the feedback, namely the sum of two delta functions, are used to quantify the effectiveness of delay distribution. We show that the distribution is effective in reducing the magnitude of the open loop transfer function, thereby, decreasing the gain-crossover frequency and improving the phase margin. Finally, inspired by Orosz et al., we demonstrate an example of how these results can be used to design a controller using delays. used to design a controller using delays.)
• Stability of Discrete Time Systems with Stochastically Delayed Feedback  + (This paper investigates the stability of l … This paper investigates the stability of linear systems with stochastic delay in discrete time. Stability of the mean and second moment of the non-deterministic system is determined by a set of deterministic discrete time equations with distributed delay. A theorem is provided that guarantees convergence of the state with convergence of the second moment, assuming the delays are identically independently distributed. The theorems are applied to a scalar equation where the stability of the equilibrium is determined.tability of the equilibrium is determined.)
• Rapid Hover to Forward Flight Transitions for a Thrust Vectored Aircraft  + (This paper investigates the use of efficie … This paper investigates the use of efficient computational algorithms for the</br>computation of a nominal trajectory for fast transition between flight modes. We use</br>differential flatness of an approximate model of the longitudinal dynamics of a thrust</br>vectored aircraft to achieve fast switching between flight modes. We investigate some</br>methods to compensate for the discrepancy between the full aerodynamic model and the</br>approximate flat model, and analyze their performance. Simulations and experimental data</br>for a thrust vectored flight control experiment at Caltech are provided to validate the</br>approach.ech are provided to validate the approach.)
• Robust Nonlinear Control Theory with Applications to Aerospace Vehicles  + (This paper is a very brief outline of an i … This paper is a very brief outline of an invited poster session giving a</br>first-year progress report on a research program with the above title being</br>carried out in the Control and Dynamical Systems (CDS) department at Caltech.</br>This 5-year grant funded by the AFOSR Partnership for Research Excellence</br>Transition (PRET) Program has a special emphasis on transitioning new methods</br>to industrial practice and thus involves a high level of industrial</br>participation. The focus of our program is fundamental research in general</br>methods of analysis and design of complex uncertain nonlinear systems, from</br>creating new mathematical theory to working to make that theory help engineers</br>solve a variety of real industrial problems. Caltech's Control and Dynamical</br>Systems department was created with precisely this goal, which is shared by our</br>industrial collaborators, led by Honeywell. Further details will be available</br>at the poster session.s will be available at the poster session.)
• On Quantized Consensus by Means of Gossip Algorithm -- Part I: Convergence Proof  + (This paper is concerned with the distribut … This paper is concerned with the distributed averaging problem subject to a quantization constraint. Given a group of agents associated with scalar numbers, it is assumed that each pair of agents can communicate with each other with a prescribed probability, and that the data being exchanged between them is quantized. In this part of the paper, it is proved that the stochastic gossip algorithm proposed in a recent paper leads to reaching the quantized consensus. Some important properties of the system in the steady-state (after reaching the consensus) are also derived. The results developed here hold true for any arbitrary quantization, provided the tuning parameter of the gossip algorithm is chosen properly. The expected value of the convergence time bounded in the second part of the paper.e bounded in the second part of the paper.)
• Quantized Consensus via Adaptive Stochastic Gossip Algorithm  + (This paper is concerned with the distribut … This paper is concerned with the distributed averaging problem over a given undirected graph. To enable every vertex to compute the average of the initial numbers sitting on the vertices of the graph, the policy is to pick an edge at random and update the values on its ending vertices based on some rules, but only in terms of the quantized data being exchanged between them. Our recent paper showed that the quantized consensus is reached under a simple updating protocol which deploys a fixed tuning factor. The current paper allows the tuning factor to be time-dependent in order to achieve two goals. First, this makes it possible to study the numerical stability of the protocol with a fixed tuning factor under a small perturbation of this parameter. Furthermore, exploiting a time-varying tuning factor facilitates the implementation of the consensus protocol and pushes the steady state of the system towards an equilibrium point, as opposed to making it oscillatory. The current paper is an important extension of our recent work, which generalizes a finite-dimensional problem toan infinite-dimensional one that is more challenging in nature.al one that is more challenging in nature.)