Property:Abstract
From Murray Wiki
Jump to navigationJump to search
This is a property of type Text.
T
Rules or specifications for autonomous vehicles are currently formulated on a case-by-case basis, and put together in a rather ad-hoc fashion. As a step towards eliminating this practice, we propose a systematic procedure for generating a set of supervisory specifications for self-driving cars that are 1) associated with a distributed assume-guarantee structure and 2) characterizable by the notion of consistency and completeness. Besides helping autonomous vehicles make better decisions on the road, the assume-guarantee contract structure also helps address the notion of blame when undesirable events occur. We give several game-theoretic examples to demonstrate applic bility of our framework. +
R
Safe and robust G&C (Guidance and Control) algorithms for onboard implementation are developed by augmenting a model predictive control technique with a safety mode. The application example herein is spacecraft small-body proximity operations where model and constraint uncertainty warrant G&C algorithms with a degree of autonomous, onboard decision capability. The algorithm enforces state and control constraints and merges two operational modes: (I) standard mode guides the spacecraft to the proximity of a target state in a robust and resolvable model-predictive manner; (II) safety mode, if activated, maintains the spacecraft near a safety reference for all time. The algorithm utilizes separate feedforward and feedback components. In standard mode, the feedforward guidance solutions come from a way-point generation algorithm that uses a discrete linear-time-varying dynamics model. This approach provides a convex formulation of the problem (solvable onboard as a second-order cone program) that includes control and state constraints; the safety-mode availability adds a constraint in this standard-mode formulation as well. The feedback guarantees standard-mode resolvability to update the guidance profile in a robust, model-predictive manner. In safety mode, an offline-designed feedforward policy with the added feedback maintains the spacecraft in a hovering state in the proximity of its position at safety-mode activation time; this provides robustness to unexpected state-constraint changes such as unexpected ground proximity during landing operations. A simulation demonstrating both the standard and safety modes is provided for a spacecraft autonomous-descent scenario toward a small asteroid with an uncertain gravity model and errors in the surface altitude constraint. +
A
Safety guarantees are built into a robust MPC (Model Predictive Control) algorithm for uncertain nonlinear systems. The algorithm is designed to obey all state and control constraints and blend two operational modes: (I) standard mode guarantees resolvability and asymptotic convergence to the origin in a robust receding-horizon manner; (II) safety mode, if activated, guarantees containment within an invariant set about a safety reference for all time. This research is motivated by physical vehicle control-algorithm design (e.g. spacecraft and hovercraft) in which operation mode changes must be considered. Incorporating safety mode provides robustness to unexpected state-constraint changes; e.g., other vehicles crossing/stopping in the feasible path, or unexpected ground proximity in landing scenarios. The safety-mode control is provided by an offline designed control policy that can be activated at any arbitrary time during standard mode. The standard-mode control consists of separate feedforward and feedback components; feedforward comes from online solution of a FHC (Finite-Horizon optimal Control problem), while feedback is designed offline to generate an invariant tube about the feedforward tra jectory. The tube provides robustness (to uncertainties and disturbances in the dynamics) and guarantees FHC resolvability. The algorithm design is demonstrated for a class of systems with uncertain nonlinear terms that have norm-bounded Jacobians. +
R
Serine integrases are bacteriophage proteins responsible for integrating the phage genome into that of the host. Synthetic biologists have co-opted these proteins into useful tools for permanent DNA logic, utilizing their specific DNA recombination abilities to build synthetic cell differentiation and genetic memory systems. Each integrase has a specific pair of DNA sequences (attP/attB sites) that it recombines, but multiple identical sites can result in unpredictable recombination. We have developed a way to control integrase activity on identical attP/attB sites by using catalytically dead Cas9 (dCas9) as a programmable binding protein that can compete with integrase for binding to specific attachment sites. Utilizing a plasmid that contains two identical Bxb1 attP sites, integration can be repressed up to 8 fold at either one of the two attP sites when guide RNA and dCas9 are present. Guide RNA sequences that bind specifically to attB, or either of two attP sites, have been developed. Future goals are to utilize this technology to construct larger and more complex integrase logic circuits. +
U
Single-cell bacterial sensors have numerous applications in human health monitoring, environmental chemical detection, and materials biosynthesis. Many previous efforts for synthetic bacteria strains seek to optimize homogenous single cell behavior. Rather than attempt to reduce noise in circuit behavior, we take advantage of heterogenous single cell responses to record sequences of chemical events within a population. Using an engineered E. coli strain with a 4-state temporal logic gate, we show, both in silico and in vivo, that stochastic digital switching within single cells results in an analog population fractionation that can be used to resolve inducer pulse duration within 30 minutes. Furthermore, these results are preserved in the genome and can be read out at a time that is much later than the time of the event. +
L
Specifications for complex engineering systems are typically decomposed into specifications for individual subsystems in a manner that ensures they are implementable and simpler to develop further. We describe a method to algorithmically construct component specifications that implement a given specification when assembled. By eliminating variables that are irrelevant to realizability of each component, we simplify the specifications and reduce the amount of information necessary for operation. We parametrize the information flow between components by introducing parameters that select whether each variable is visible to a component or not. The decomposition algorithm identifies which variables can be hidden while pre- serving realizability and ensuring correct composition, and these are eliminated from component specifications by quantification and conversion of binary decision diagrams to formulas. The resulting specifications describe component viewpoints with full information with respect to the remaining variables, which is essential for tractable algorithmic synthesis of implementations. The specifications are written in TLA+, with liveness properties restricted to an implication of conjoined recurrence properties, known as GR(1). We define an operator for forming open systems from closed systems, based on a variant of the “while-plus” operator. This operator simplifies the writing of specifications that are realizable without being vacuous. To convert the generated specifications from binary decision diagrams to readable formulas over integer variables, we symbolically solve a minimal covering problem. We show with examples how the method can be applied to obtain contracts that formalize the hierarchical structure of system design. +
C
Standard schemes in system identification and adaptive control rely on persistence of excitation to guaran- tee parameter convergence. Inspired by networked systems, we extend parameter adaptation to the multi-agent setting by combining a gradient law with consensus dynamics. The gradient law introduces a learning signal, while consensus dynamics preferentially push each agentâs parameter estimates toward those of its neighbors. We show that the resulting online, decentralized parameter estimator combines local and neighboring information to identify the true parameters even if no single agent employs a persistently exciting input. We also elaborate upon collective persistence of excitation in networked adaptive algorithms. +
Station keeping and reorientation control of a cluster of fully-actuated
low-thrust micro-satellites is considered in this paper. We address the control
problem by taking advantage of the fully-actuated structure of the micro-satellite.
We propose a very general open-loop solution by solving in real-time constrained
trajectory generation problems for stationkeeping and reorientation. Performance
of this methodology is reported for a typical micro-satellite format ion flying
space mission using the Nonlinear Trajectory Generation software package. +
A
A Bayesian approach to inferring chemical signal timing and amplitude in a temporal logic gate using the cell population distributional response +
Stochastic gene expression poses an important challenge for engineering robust behaviors in a heterogeneous cell population. Cells address this challenge by operating on distributions of cellular responses generated by noisy processes. Similarly, a previously published temporal logic gate considers the distribution of responses across a cell population under chemical inducer pulsing events. The design uses a system of two integrases to engineer an E. coli strain with four DNA states that records the temporal order of two chemical signal events. The heterogeneous cell population response was used to infer the timing and duration of the two chemical signals for a small set of events. Here we use the temporal logic gate system to address the problem of extracting information about chemical signal events. We use the heterogeneous cell population response to infer whether any event has occurred or not and also to infer its properties such as timing and amplitude. Bayesian inference provides a natural framework to answer our questions about chemical signal occurrence, timing, and amplitude. We develop a probabilistic model that incorporates uncertainty in the how well our model captures the cell population and in how well a sample of measured cells represents the entire population. Using our probabilistic model and cell population measurements taken every five minutes on generated data, we ask how likely it was to observe the data for parameter values that describe square-shaped inducer pulses. We compare the likelihood functions associated with the probabilistic models for the event with the chemical signal pulses turned on versus turned off. Hence, we can determine whether an event of chemical induction of integrase expression has occurred or not. Using Markov Chain Monte Carlo, we sample the posterior distribution of chemical pulse parameters to identify likely pulses that produce the data measurements. We implement this method and obtain accurate results for detecting chemical inducer pulse timing, length, and amplitude. We can detect and identify chemical inducer pulses as short as half an hour, as well as all pulse amplitudes that fall under biologically relevant conditions.
A stochastic framework for the design of transient and steady state behavior of biochemical reaction networks +
Stochasticity plays an essential role in biochemical systems. Stochastic behaviors of bimodality, excitability, and fluctuations have been observed in biochemical reaction networks at low molecular numbers. Stochastic dynamics can be captured by modeling the system using a forward Kolmogorov equation known in the biochemical literature as the chemical master equation. The chemical master equation describes the time evolution of the probability distributions of the molecule species. We develop a stochastic framework for the design of these time evolving probability distributions that includes specifying their uni-/multi-modality, their first moments, and their rate of convergence to the stationary distribution. By solving the corresponding optimizations programs, we determine the reaction rates of the biochemical systems that satisfy our design specifications. We then apply the design framework to examples of biochemical reaction networks to illustrate its strengths and limitations. +
Substantial reductions in aircraft
size are possible if shorter, more aggressive, serpentine inlet ducts are used
for low-observability constrained propulsion installations. To obtain this benefit,
both inlet separation and compressor stall dynamics must be controlled. In this
paper the integrated control of this coupled inlet/compression system is considered.
Initial results are shown using separation point actuation to control both separation
and stall dynamics. Calculations show that separation can be substantially reduced
with approximately 1.2% core flow, based on scaling previous results. Simulation
results using a medium fidelity model show that proportional control of distortion
has little effect on stall behavior. +
S
Successful high-speed autonomous navigation requires integration of tools from robotics, control theory, computer vision, and systems engineering. This thesis presents work that develops and combines these tools in the context of navigating desert terrain.
A comparative analysis of reactive, behavior-based, and deliberative control architectures provides important guidelines for design of robotic systems. These guidelines depend on the particular task and environment of the vehicle. Two important factors are identified which guide an effective choice between these architectures: dynamic feasibility for the vehicle, and predictability of the environment. This is demonstrated by parallels to control theory, illustrative examples, simulations, and analysis of Bob and Alice---Caltech's full-scale autonomous ground vehicle entries in the 2004 and 2005 Grand Challenge races, respectively.
Further, new model-based methods are developed for constructing and maintaining estimates of terrain elevation and road geometry. These are demonstrated in simulation and in fully autonomous operation of Alice, including accurate detection and tracking of the centerline of desert roads at speeds up to 5 m/s. Finally, Alice's navigation architecture is presented in full along with experimental results that demonstrate its capabilities. +
C
Synthetic biologists have turned towards quorum systems as a path for building sophisticated microbial consortia that exhibit group decision making. Currently, however, even the most complex consortium circuits rely on only one or two quorum sensing systems, greatly restricting the available design space. High-throughput characterization of available quorum sensing systems is useful for finding compatible sets of systems that are suitable for a defined circuit architecture. Recently, cell-free systems have gained popularity as a test-bed for rapid prototyping of genetic circuitry. We take advantage of the transcription-translation cell-free system to characterize three commonly used Lux-type quorum activators, Lux, Las, and Rpa. We then compare the cell-free characterization to results obtained in vivo. We find significant genetic crosstalk in both the Las and Rpa systems and substantial signal crosstalk in Lux activation. We show that cell-free characterization predicts crosstalk observed in vivo. +
B
Synthetic gene expression is highly sensitive to intragenic compositional context (promoter structure, spacing regions between promoter and coding sequences, and ribosome binding sites). However, much less is known about the effects of intergenic compositional context (spatial arrangement and orientation of entire genes on DNA) on expression levels in synthetic gene networks. We compare expression of induced genes arranged in convergent, divergent, or tandem orientations. Induction of convergent genes yielded up to 400% higher expression, greater ultrasensitivity, and dynamic range than divergent- or tandem-oriented genes. Orientation affects gene expression whether one or both genes are induced. We postulate that transcriptional interference in divergent and tandem genes, mediated by supercoiling, can explain differences in expression and validate this hypothesis through modeling and in vitro supercoiling relaxation experiments. Treatment with gyrase abrogated intergenic context effects, bringing expression levels within 30% of each other. We rebuilt the toggle switch with convergent genes, taking advantage of supercoiling effects to improve threshold detection and switch stability. +
D
Data-driven network models for genetic circuits from time-series data with incomplete measurements +
Synthetic gene networks are frequently conceptualized and visualized as static graphs. This view of biological programming stands in stark contrast to the transient nature of biomolecular interaction, which is frequently enacted by labile molecules that are often unmeasured. Thus, the network topology and dynamics of synthetic gene networks can be difficult to verify in vivo or in vitro, due to the presence of unmeasured biological states. Here we introduce the dynamical structure function as a new mesoscopic, data-driven class of models to describe gene networks with incomplete measurements. We introduce a network reconstruction algorithm and a code base for reconstructing the dynamical structure function from data, to enable discovery and visualization of graphical relationships in a genetic circuit diagram as time-dependent functions rather than static, unknown weights. We prove a theorem, showing that dynamical structure functions can provide a data-driven estimate of the size of crosstalk fluctuations from an idealized model. We illustrate this idea with numerical examples. Finally, we show how data-driven estimation of dynamical structure functions can explain failure modes in two experimentally implemented genetic circuits, a historical genetic circuit and a new E. coli based transcriptional event detector. +
M
Modeling predicts that CRISPR-based activators, unlike CRISPR-based repressors, scale well with increasing gRNA competition and dCas9 bottlenecking +
Synthetic transcriptional networks built from CRISPR-based repressors (CRISPRi) rely on shared use of a core dCas9 protein. In E. coli, CRISPRi cannot support more than about a dozen simultaneous gRNAs before the fold repression of any individual gRNA drops below 10x. We show with a simple model based on previous characterization of competition in CRISPRi that activation by CRISPR-based activators (CRISPRa) is much less sensitive to dCas9 bottle-necking than CRISPRi. We predict that E. coli should be able to support dozens to hundreds of CRISPRa gRNAs at >10-fold activation. +
Targeted transcriptional repression with catalytically inactive Cas9 (CRISPRi) can be used to build gene regulatory nets similar in principle to those made with traditional transcription factors, and promises to do so with better orthogonality, programmability, and extensibility. We use a simple dynamical model of CRISPRi to understand its behavior and requirements, and to show that CRISPRi can recapitulate several classic gene regulatory circuits, including the repressilator, a toggle switch, and an incoherent feed-forward loop pulse generator. Our model also predicts that these circuits are highly sensitive to promoter leak, but that promoter leak can be offset with active degradation of dCas. We provide specifications for required fold-repression and dCas degradation rates for several dynamic circuits. Our modeling reveals key engineering requirements and considerations for the construction of dynamic CRISPRi circuits, and provides a roadmap for building those circuits. +
N
Temporal dynamics in many biomolecular circuits can change with temperature because of the temperature dependence of underlying reaction rate parameters. It is generally unclear what circuit mechanisms can inherently facilitate robustness in the dynamics to variations in temperature. Here, we address this issue using a combination of mathematical models and experimental measure- ments in a cell-free transcription-translation system. We find that negative transcriptional feedback can reduce the eâµect of temperature variation on circuit dynamics. Further, we find that effective negative feedback due to first-order degradation mechanisms can also enable such a temperature robustness effect. Finally, we estimate temperature dependence of key parameters mediating such negative feedback mechanisms. These results should be useful in the design of temperature robust circuit dynamics. +
D
Temporal logic based synthesis approaches are often used to find trajectories that are correct-by-construction in systems–eg. synchronization for multi-agent hybrid systems, reactive motion planning for robots. However, the scalability of such approaches is of concern and at times a bottleneck when transitioning from theory to practice. In this paper, we identify a class of problems in the GR(1) fragment of linear-time temporal logic (LTL) where the synthesis problem allows for a decomposition that enables easy parallelization. This decomposition also reduces the alternation depth, resulting in more efficient synthesis. A multi-agent robot gridworld example with coordination tasks is presented to demonstrate the application of the developed ideas and also to perform empirical analysis for benchmarking the decomposition-based synthesis approach. +
O
Temporal logics have proven effective for correct-by-construction
synthesis of controllers for a wide range of
applications. Receding horizon frameworks mitigate the
computational intractability of reactive synthesis for temporal
logic, but have thus far been limited by pursuing a single
sequence of short horizon problems to the current goal. We
propose a receding horizon algorithm for reactive synthesis that
automatically determines a path to the currently pursued goal at
runtime, in response to a nondeterministic environment. This is
achieved by allowing each short horizon to have multiple local
goals, and determining which local goal to pursue based on the
current global goal, currently perceived environment and a
pre-computed invariant dependent on each global goal. We
demonstrate the utility of this additional flexibility in
grant-response tasks, using a search-and-rescue
example. Moreover, we show that these goal-dependent invariants
mitigate the conservativeness of the receding horizon approach. +
