Hybrid vehicle techniques have been widely studied recently because of their potential to significantly improve the fuel economy and drivability future ground vehicles. Due dual-power-source nature these vehicles, control strategies based on engineering intuition frequently fail fully explore advanced In this paper, we present a procedure for design near-optimal power management strategy. The starts by defining cost function, such as minimizing combination consumption selected emission...

Planar, underactuated, biped walkers form an important domain of applications for hybrid dynamical systems. This paper presents the design exponentially stable walking controllers general planar bipedal systems that have one degree-of-freedom greater than number available actuators. The within-step control action creates attracting invariant set - a two-dimensional zero dynamics submanifold full model $whose restriction admits scalar linear time-invariant return map. Exponentially periodic...

This paper develops a control methodology that unifies barrier functions and Lyapunov through quadratic programs. The result is demonstrated on adaptive cruise control, which presents both safety performance considerations, as well actuator bounds. We begin by presenting novel notion of function associated with set, formulated in the context Lyapunov-like conditions; existence satisfying these conditions implies forward invariance set. formulation naturally yields (CBF), yielding inequality...

Safety critical systems involve the tight coupling between potentially conflicting control objectives and safety constraints. As a means of creating formal framework for controlling this form, with view toward automotive applications, paper develops methodology that allows conditions-expressed as barrier functions-to be unified performance objectives-expressed Lyapunov functions-in context real-time optimization-based controllers. conditions are specified in terms forward invariance set,...

Describes the design, construction and control of an experimental bipedal robot platform for study walking.

The supervisory control strategy of a hybrid vehicle coordinates the operation sub-systems to achieve performance targets such as maximizing fuel economy and reducing exhaust emissions. This high-level problem is commonly referred power management problem. In past, many strategies were developed on basis few pre-defined driving cycles, using intuition heuristics. resulting was often inherently cycle-beating lacked guaranteed level optimality. this study, tackled from stochastic viewpoint. An...

This paper addresses the problem of exponentially stabilizing periodic orbits in a special class hybrid models-systems with impulse effects-through control Lyapunov functions. The orbit is assumed to lie C <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> submanifold Z that contained zero set an output function and invariant under both continuous discrete dynamics; associated restriction dynamics are termed dynamics. furthermore be stable...

The planar bipedal testbed MABEL contains springs in its drivetrain for the purpose of enhancing both energy efficiency and agility dynamic locomotion. While potential energetic benefits are well documented literature, feedback control designs that effectively realize this lacking. In paper, we extend apply methods virtual constraints hybrid zero dynamics, originally developed rigid robots with a single degree underactuation, to MABEL, walker novel compliant transmission multiple degrees...

Barrier functions (also called certificates) have been an important tool for the verification of hybrid systems, and also played roles in optimization multi-objective control. The extension a barrier function to controlled system results control function. This can be thought as being analogous how Sontag extended Lyapunov order enable controller synthesis stabilization tasks. A enables safety requirements specified by forward invariance set using Lyapunov-like condition. paper develops...

This paper presents three feedback controllers that achieve an asymptotically stable, periodic, and fast walking gait for a 3D (spatial) bipedal robot consisting of torso, two legs, passive (unactuated) point feet. The contact between the surface is assumed to inhibit yaw rotation. studied has 8 DOF in single support phase 6 actuators. interest studying robots with feet robot's natural dynamics must be explicitly taken into account balance while walking. We use extension method virtual...

A hybrid controller that induces provably stable running gaits on an asymmetric spring loaded inverted pendulum (ASLIP) is developed. The acts two levels. On the first level, continuous within-stride control asymptotically imposes a (virtual) holonomic constraint corresponding to desired torso posture, and creates invariant surface which two-degree-of-freedom restriction dynamics of closed-loop system (i.e., zero dynamics) diffeomorphic center-of-mass (SLIP). second event-based stabilizes...

Legged robots require knowledge of pose and velocity in order to maintain stability execute walking paths. Current solutions either rely on vision data, which is susceptible environmental lighting conditions, or fusion kinematic contact data with measurements from an inertial measurement unit (IMU). In this work, we develop a contact-aided invariant extended Kalman filter (InEKF) using the theory Lie groups observer design. This combines contact-inertial dynamics forward corrections estimate...

The Cassie bipedal robot designed by Agility Robotics is providing academics with a common platform for sharing and comparing algorithms locomotion, perception, navigation. This paper focuses on feedback control standing walking using the methods of virtual constraints gait libraries. controller was implemented six weeks after arrived at University Michigan allowed it to stand in place as well walk over sidewalks, grass, snow, sand, burning brush. also enables ride Segway. Software...

This paper focuses on the development of asymptotic observers for nonlinear discrete-time systems. It is argued that instead trying to imitate linear observer theory, problem constructing a can be more fruitfully studied in context solving simultaneous equations. In particular, it shown discrete Newton method, properly interpreted, yields an large class systems, while continuous method may employed obtain global observer. Furthermore, analyzed how use Broyden's structure affects observer's...

The convergence aspects of the extended Kalman filter, when used as a deterministic observer for nonlinear discrete-time system, are analyzed. case systems with output maps well linear is treated and conditions needed to ensure uniform boundedness certain Riccati equations related observability properties underlying system. Furthermore, we show filter without any priori assumptions on error covariances long states stay within convex compact domain.

The primary goal of this paper is to demonstrate a means prove asymptotically stable walking in an underactuated, planar, five-link biped robot model. analysis assumes rigid contact model when the swing leg impacts ground and instantaneous double support phase. specific analyzed corresponds prototype under development by Centre National de la Recherche Scientifique (CNRS), Paris, France. A secondary establish viability theoretically motivated control law. This explored number ways. First, it...

A linear algebraic framework for the analysis of rank properties nonlinear systems is introduced. This gives a high-level interpretation several existing algorithms built around recursive computation certain ranks associated with right-invertibility, left-invertibility, and dynamic decoupling. Furthermore, it can be used to establish links between these differential approach, as well solve some static noninteracting control problems.

Hybrid vehicle fuel economy performance is highly sensitive to the energy management strategy used regulate power flow among various sources and sinks. Optimal non-causal solutions are easy determine if drive cycle known a priori. It very challenging design causal controllers that yield good for range of possible driver behavior. Additional challenges come in form constraints on powertrain activity, such as shifting starting engine, which commonly called "drivability" metrics can adversely...

This paper presents a novel method to address the actuator saturation for nonlinear hybrid systems by directly incorporating user-defined input bounds in controller design. In particular, we consider application of bipedal walking and show that our [based on quadratic programming (QP) implementation control Lyapunov function (CLF)-based controller] enables gradual performance degradation while still continuing walk under increasingly stringent bounds. We draw previous work, which has...

Motivated by the problem of controlling walking in a biped with series compliant actuation, this paper develops two main theorems relating to stabilization periodic orbits systems impulse effects. The first result shows that when orbit system effects lies within hybrid invariant manifold, there exist local coordinate transforms under which Jacobian linearization Poincare return map has block upper triangular structure. One diagonal is as restricted also called zero dynamics. other product...

This paper develops a control approach with correctness guarantees for the simultaneous operation of lane keeping and adaptive cruise control. The safety specifications these driver assistance modules are expressed in terms set invariance. Control barrier functions (CBFs) used to design family solutions that guarantee forward invariance set, which implies satisfaction specifications. CBFs synthesized through combination sum-of-squares program physics-based modeling optimization. A real-time...

This paper presents an obstacle avoidance algorithm for low speed autonomous vehicles (AV), with guaranteed safety. A supervisory control is constructed based on a barrier function method, which works in plug-and-play fashion any lower level navigation algorithm. When the risk of collision low, not active; when high, distance to "avoidable set," controller will intervene, using mixed integer program ensure safety minimal effort. method applied solve and pedestrian problem AV. Its performance...