Biblio

This paper documents and discusses the design of a low-cost Flapping-Wing Micro Air Vehicle (FW-MAV) designed to be easy to fabricate using readily available materials and equipment. Basic theory of operation as well as the rationale underlying various design decisions will be provided. Using this paper, it should be possible for readers to construct their own devices quickly and at little expense.

Previously, we introduced Evolutionary Model Consistency Checking (EMCC) as an adjunct to Evolvable and Adaptive Hardware (EAH) methods. The core idea was to dual-purpose objective function evaluations to simultaneously enable EA search of hardware configurations while simultaneously enabling a model-based inference of the nature of the damage that necessitated the hardware adaptation. We demonstrated the efficacy of this method by modifying a pair of EAH oscillators inside a simulated Flapping-Wing Micro Air Vehicle (FW-MAV). In that work, we were able to show that one could, while online in normal service, evolve wing gait patterns that corrected altitude control errors cause by mechanical wing damage while simultaneously determining, with high precision, what the wing lift force deficits that necessitated the adaptation. In this work, we extend the method to be able to also determine wing drag force deficits. Further, we infer the now extended set of four unknown damage estimates without substantially increasing the number of objective function evaluations required. In this paper we will provide the outlines of a formal derivation of the new inference method plus experimental validation of efficacy. The paper will conclude with commentary on several practical issues, including better containment of estimation error by introducing more in-flight learning trials and why one might argue that these techniques could eventually be used on a true free-flying flapping wing vehicle.

Those employing Evolutionary Algorithms (EA) are constantly challenged to engineer candidate solution representations that balance expressive power (I.E. can a wide variety of potentially useful solutions be represented?) and meta-heuristic search support (I.E. does the representation support fast acquisition and subsequent fine-tuning of adequate solution candidates). In previous work with a simulated insect-like Flapping-Wing Micro Air Vehicle (FW-MAV), an evolutionary algorithm was employed to blend descriptions of wing flapping patterns to restore correct flight behavior after physical damage to one or both of the wings. Some preliminary work had been done to reduce the overall size of the search space as a means of improving time required to acquire a solution. This of course would likely sacrifice breadth of solutions types and potential expressive power of the representation. In this work, we focus on methods to improve performance by augmenting EA search to dynamically restrict and open access to the whole space to improve solution acquisition time without sacrificing expressive power of the representation. This paper will describe some potential restriction/access control methods and provide preliminary experimental results on the efficacy of these methods in the context of adapting FW-MAV wing gaits.

This paper studies the multi-agent average consensus problem under the requirement of differential privacy of the agents' initial states against an adversary that has access to all messages. As a fundamental limitation, we first establish that a differentially private consensus algorithm cannot guarantee convergence of the agents' states to the exact average in distribution, which in turn implies the same impossibility for other stronger notions of convergence. This result motives our design of a novel differentially private Laplacian consensus algorithm in which agents linearly perturb their state-transition and message-generating functions with exponentially decaying Laplace noise. We prove that our algorithm converges almost surely to an unbiased estimate of the average of the agents' initial states, compute the exponential mean-square rate of convergence, and formally characterize its differential privacy properties. Furthermore, we also find explicit optimal values of the design parameters that minimize the variance of the algorithm's convergence point around the exact average. Various simulations illustrate our results.

We study a class of distributed convex constrained optimization problem where a group of agents aims to minimize the sum of individual objective functions while each desires to keep its function differentially private. We prove the impossibility of achieving differential privacy using strategies based on perturbing with noise the inter-agent messages when the underlying noise-free dynamics is asymptotically stable. This justifies our algorithmic solution based on the perturbation of the individual objective functions with Laplace noise within the framework of functional differential privacy. We carefully design post-processing steps that ensure the perturbed functions regain the smoothness and convexity properties of the original functions while preserving the differentially private guarantees of the functional perturbation step. This methodology allows to use any distributed coordination algorithm to solve the optimization problem on the noisy functions. Finally, we explicitly bound the magnitude of the expected distance between the perturbed and true optimizers, and characterize the privacy-accuracy trade-off. Simulations illustrate our results.

This paper studies the problem of privacy-preserving average consensus in multi-agent systems. The network objective is to compute the average of the initial agent states while keeping these values differentially private against an adversary that has access to all inter-agent messages. We establish an impossibility result that shows that exact average consensus cannot be achieved by any algorithm that preserves differential privacy. This result motives our design of a differentially private discrete-time distributed algorithm that corrupts messages with Laplacian noise and is guaranteed to achieve average consensus in expectation. We examine how to optimally select the noise parameters in order to minimize the variance of the network convergence point for a desired level of privacy.

This paper proposes a novel distributed event-triggered algorithmic solution to the multi-agent average consensus problem for networks whose communication topology is described by weight-balanced, strongly connected digraphs. The proposed event-triggered communication and control strategy does not rely on individual agents having continuous or periodic access to information about the state of their neighbors. In addition, it does not require the agents to have a priori knowledge of any global parameter to execute the algorithm. We show that, under the proposed law, events cannot be triggered an infinite number of times in any finite period (i.e., no Zeno behavior), and that the resulting network executions provably converge to the average of the initial agents' states exponentially fast. We also provide weaker conditions on connectivity under which convergence is guaranteed when the communication topology is switching. Finally, we also propose and analyze a periodic implementation of our algorithm where the relevant triggering functions do not need to be evaluated continuously. Simulations illustrate our results and provide comparisons with other existing algorithms.