Self-driving vehicles are a maturing technology with the potential to reshape mobility by enhancing safety, accessibility, efficiency, and convenience of automotive transportation. Safety-critical tasks that must be executed self-driving vehicle include planning motions through dynamic environment shared other pedestrians, their robust executions via feedback control. The objective this paper is survey current state art on control algorithms particular regard urban setting. A selection...

This paper develops nonsmooth Lyapunov stability theory and LaSalle's invariance principle for a class of Lipschitz continuous functions absolutely state trajectories. Computable tests based on Filipov's differential inclusion Clarke's generalized gradient are derived. The primary use these results is in analyzing the equilibria equations with discontinuous right-hand side such as dynamic systems or variable structure control.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML"...

An inversion procedure is introduced for nonlinear systems which constructs a bounded input trajectory in the preimage of desired output trajectory. In case minimum phase systems, produced agrees with that generated by Hirschorn's inverse dynamic system; however, noncausal (rather than unstable) nonminimum case. addition, analysis leads to simple geometric connection between unstable manifold system zero dynamics and noncausality With addition stabilizing feedback trajectory, asymptotically...

This paper develops a calculus for computing Filippov's differential inclusion which simplifies the analysis of dynamical systems described by equations with discontinuous right-hand side. In particular, when slightly generalized Lyapunov theory is used, rigorous stability variable structure routine. As an example, control law rigid-link robot manipulators and its proved.

The authors investigate the effects of state disturbances, output noise, and errors in initial conditions on a class learning control algorithms. They present simple algorithm exhibit, via concise proof, bounds asymptotic trajectory for learned input corresponding trajectories. Furthermore, these are continuous functions condition errors, zero absence disturbances.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Repetitive controllers are generally applied to reject periodic disturbances and track reference signals with a known period. Their design is based on the internal model principle, proposed by Francis Wonham (1975). This paper describes new finite-dimensional SISO repetitive controller for two different classes of nonlinear plants. Simulation results show asymptotic tracking signal in closed loop up Nth harmonic frequency. A proof robustness control system small nonlinearities, like actuator...

We present a new scheme for the adaptive control of mechanical manipulators along with proof convergence. This work is an extension our earlier [Craig, Hsu and Sastry] [1]. The does not require measurement joint accelerations needs less computation. illustrate theory some simulations.

We extend stable inversion to nonlinear time-varying systems and study computational issues-the technique is applicable minimum-phase as well nonminimum-phase systems. The new, even in the linear case, relies on partitioning (the dichotomic split of) linearized system dynamics into time-varying, stable, unstable, submanifolds. This used build filters which are, turn, basis of a contraction find bounded inverse input-state trajectory. Finding trajectory allows development exact-output...

An adaptive linearization scheme for torque-ripple cancellation is presented, and the stability robustness are established. By taking a new approach in parameterizing motor dynamics, number of adapted parameters reduced by factor two relative to standard approach. This parameterization unique periodic property enable authors find conditions on exogenous signals which guarantee persistency excitation. The develop result which, roughly speaking, shows that allowable model perturbation does not...

In this paper we describe a new tracking control law for multijoint flexible-link manipulators. The scheme is synthesis of the inverse dynamics solution flexible manipulators developed by Bayo at UCSB and theory rigid-link put forth Bayard, Wen others. We show that passive joint controllers, together with feedforward nominal torques corresponding to desired end-effector trajectory, results in exponentially stable control. Stability proved (local trajectory) large class controllers which...

The modeling of torque-ripple in hybrid step motors and its cancellation using adaptive linearization control are discussed. Although the nonlinear this problem can fit into a general framework, representation which reduces number adapted parameters per harmonic by half is used. By doing so, it possible to prove conditions on exogenous signals guarantee persistency excitation regressor, hence exponential stability unperturbed system. It shown that system robust class state-...

An important control problem is output tracking. One approach, for slightly nonminimum phase systems to approximate the system by a minimum one. Another approach track signals generated an exosystem. These methods concentrate on properties of IO without any attempt investigate and use structural IS system. In this paper authors propose alternative point view based following simple remark: if one had state tracker (i.e. controller able reference trajectory), tracking would reduce generation...

Learning control is a very effective approach for tracking in processes occurring repetitively over fixed interval of time. In this paper, an iterative learning (ILC) algorithm proposed to accommodate general class nonlinear, nonminimum-phase plants with disturbances and initialization errors. The requires the computation approximate inverse linearized plant rather than exact inverse. An advantage that output need not be differentiated. A bound on asymptotic trajectory error exhibited via...

Solves the exact output tracking problem for nonlinear time-varying systems. The technique is even applicable in nonminimum phase case, and uses an inverse trajectory feedforward. feedforward coupled with a more conventional feedback control law stable tracking. inversion new, linear relies on partitioning linearized system dynamics into unstable manifolds. This so-called dichotomic split used to build filters which, turn, are basis of contraction find trajectory. method local trajectory,...

Abstract A trajectory based on a Gaussian velocity profile is presented as an alternative to the double square pulse acceleration profile. The new leads fast positioning of tip flexible robot with minimal excitation high‐frequency modes. torques necessary move according this show very smooth behavior. absence content, present when accelerations are considered, eliminates Occurrence undesired residual vibrations produced by modeling uncemnties at high frequencies. excellent results obtained...

A stability analysis and design method for MEMS resonators is presented. The frequency characteristics of a laterally vibrating resonator are analyzed. With the fabrication error on sidewall structure being considered, first second order sensitivities to derived. simple relationship between proof mass area perimeter, beam width, developed single material structures, which expresses that perimeter times width should equal six mass. Design examples given multi-layer structures. results...

The robustness of a class learning control algorithms to state disturbances, output noise, and errors in initial conditions is studied. A simple algorithm presented, bounds on the asymptotic trajectory for learned input corresponding trajectories are exhibited. These continuous functions condition errors, disturbance, zero absence these disturbances.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

A simple and efficient algorithm is developed to generate a 2"-tree (generalized quad-tree) representation of the free configuration space for manipulator moving in workspace with obstacles. The based on existence uniform bounds Jacobians relating differential motions points joint motions. This can be searched collision-free path optimized dynamic constraints produce an executable trajectory.

A nonlinear phase estimator for a phase-shift keyed (PSK)-modulated carrier has been developed by Viterbi and Viterbi. Their analysis is extended, an optimal or "matched" nonlinearity derived the estimator.

A novel electromechanical valve actuation system comprised of a linear actuator, valve, and energy storing cam/spring mechanism is presented. The dynamics are modeled using Lagrangian mechanics, minimum-energy point-to-point optimal control problem solved to find an trajectory input. input used as feedforward component in transition controller move the between open closed positions. Between transitions, simple stabilizes high-order model capturing distributed nature springs validate state...

For over one hundred years, the principle of continuously and infinitely variable transmissions (C/IVT) has eluded practical automotive utility. With advent durable materials, sophisticated electronic controls, improved lubricants, a viable assortment C/IVT powertrain systems with innovative ratio shift control strategies is now feasible. A few these progressive system designs are described in this survey. Pertinent fundamental relationships powertrains also discussed.

Kharitonov's theorem is used to derive a robust stability condition for proportional-integral-derivative (PID)-controlled multi-degree-of-freedom mechanical systems. The characteristic equation of such system given as determinant third-order polynomial with matrix coefficients from which scalar interval obtained. A simple procedure designing PID controllers these systems described, and new Kharitonov-like result states roughly that controller designed an upper bounding inertia results in...

An algorithm is presented for generating near-time-optimal trajectories an open-kinematic-chain manipulator moving in a cluttered workspace. This the first to guarantee bounds on closeness of approximation time-optimal trajectory. The running time and space required are polynomial desired accuracy approximation. user may also specify tolerances by which must clear obstacles workspace allow modeling control errors.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML"...