John P. Swensen

Kernel-based visual servoing

Traditionally, visual servoing is separated into tracking and control subsystems. This separation, though convenient, is not necessarily well justified. When tracking and control strategies are designed independently, it is not clear how to optimize them to achieve a certain task. In this work, we propose a framework in which spatial sampling kernels - borrowed from the tracking and registration literature - are used to design feedback controllers for visual servoing. The use of spatial sampling kernels provides natural hooks for Lyapunov theory, thus unifying tracking and control and providing a framework for optimizing a particular servoing task. As a first step, we develop kernel-based visual servos for a subset of relative motions between camera and target scene. The subset of motions we consider are 2D translation, scale, and roll of the target relative to the camera. Our approach provides formal guarantees on the convergence/stability of visual servoing algorithms under putatively generic conditions.

Closed-Form Differential Kinematics for Concentric-Tube Continuum Robots with Application to Visual Servoing

Active cannulas, so named because of their potential medical applications, are a new class of continuum robots consisting of precurved, telescoping, elastic tubes. As individual component tubes are actuated at the base relative to one another, an active cannula changes shape to minimize stored elastic energy. Here, we derive the differential kinematics of a general n-tube active cannula while accounting for torsional compliance. We experimentally validate the Jacobian using a three-link prototype in a simple stereo visual servoing scheme.

Printing Three-Dimensional Electrical Traces in Additive Manufactured Parts for Injection of Low Melting Temperature Metals

While techniques exist for the rapid prototyping of mechanical and electrical components separately, this paper describes a method where commercial additive manufacturing (AM) techniques can be used to concurrently construct the mechanical structure and electronic circuits in a robotic or mechatronic system. The technique involves printing hollow channels within 3D printed parts that are then filled with a low melting point liquid metal alloy that solidifies to form electrical traces. This method is compatible with most conventional fused deposition modeling and stereolithography (SLA) machines and requires no modification to an existing printer, though the technique could easily be incorporated into multimaterial machines. Three primary considerations are explored using a commercial fused deposition manufacturing (FDM) process as a testbed: material and manufacturing process parameters, simplified injection fluid mechanics, and automatic part generation using standard printed circuit board (PCB) software tools. Example parts demonstrate the ability to embed circuits into a 3D printed structure and populate the surface with discrete electronic components. [DOI: 10.1115/1.4029435]

Demonstration of the James Webb Space Telescope commissioning on the JWST testbed telescope

The one-meter Testbed Telescope (TBT) has been developed at Ball Aerospace to facilitate the design and implementation of the wavefront sensing and control (WFS&C) capabilities of the James Webb Space Telescope (JWST). The TBT is used to develop and verify the WFS&C algorithms, check the communication interfaces, validate the WFS&C optical components and actuators, and provide risk reduction opportunities for test approaches for later full-scale cryogenic vacuum testing of the observatory. In addition, the TBT provides a vital opportunity to demonstrate the entire WFS&C commissioning process. This paper describes recent WFS&C commissioning experiments that have been performed on the TBT.

Torsional Dynamics of Steerable Needles: Modeling and Fluoroscopic Guidance

Needle insertions underlie a diversity of medical interventions. Steerable needles provide a means by which to enhance existing needle-based interventions and facilitate new ones. Tip-steerable needles follow a curved path and can be steered by twisting the needle base during insertion, but this twisting excites torsional dynamics that introduce a discrepancy between the base and tip twist angles. Here, we model the torsional dynamics of a flexible rod-such as a tip-steerable needle-during subsurface insertion and develop a new controller based on the model. The torsional model incorporates time-varying mode shapes to capture the changing boundary conditions inherent during insertion. Numerical simulations and physical experiments using two distinct setups-stereo camera feedback in semitransparent artificial tissue and feedback control with real-time X-ray imaging in optically opaque artificial tissue-demonstrate the need to account for torsional dynamics in control of the needle tip.

Torsional dynamics compensation enhances robotic control of tip-steerable needles

Needle insertions serve a critical role in a wide variety of medical interventions. Steerable needles provide a means by which to enhance existing percutaneous procedures and afford the development of entirely new ones. Here, we present a new time-varying model for the torsional dynamics of a steerable needle, along with a new controller that takes advantage of the model. The torsional model incorporates time-varying mode shapes to capture the changing boundary conditions caused during insertion of the needle into the tissue. Extensive simulations demonstrate the improvement over a model that neglects torsional dynamics and illustrates the possible effect of torsional model order on efficacy. Pilot feedback control experiments, conducted in artificial tissue (plastisol) under stereo image guidance, validate the overall approach: our results substantially out-perform previously reported experimental results on controlling tip-steerable needles.

An almost global estimator on SO(3) with measurement on S 2

This paper presents an almost globally convergent state estimator for the orientation of a rotating rigid body. The estimator requires knowledge of the angular velocity of the body at each time instant and the measurement consists of a single unit vector on the body, which we take without loss of generality to be the first column of the rotation matrix. The stability proof involves a relatively simple Lyapunov and invariance-like analysis. A mild non-degeneracy constraint on the control input guarantees the fulfillment of the invariance criterion. We apply the result to needle-tip orientation estimation for tip-steerable needles.

Simple, scalable active cells for articulated robot structures

The proposed research effort explores the development of active cells - simple contractile electromechanical units that can be used as the material basis for larger articulable structures. Each cell, which might be considered a “muscle unit”, consists of a contractile Nitinol SMA core with conductive terminals. Large numbers of these cells might be combined and externally powered to change phase, contracting to either articulate with a large strain or increase the stiffness of the ensemble, depending on the cell design. Unlike traditional work in modular robotics, the approach presented here focuses on cells that have a simplistic design and function, are inexpensive to fabricate, and are eventually scalable to sub-millimeter sizes, working towards our vision of robot structures that can be custom-fabricated from large numbers of general cell units, similar to biological structures.

The connectedness of packed circles and spheres with application to conductive cellular materials.

In this paper, we examine the static connectivity of 2D and 3D arrays of spherical cells with conductive paths, and the associated power dissipation in the individual cells. Herein, we use the term “cellular material” to describe the ensemble of many cells, in contrast to the more traditional use of the term for foams and honeycomb materials. Using a numerical analytical approach from highly parallel resistor arrays, we examine the cells and ensemble structures in terms of their connectivity, defined as the number of cells that are dissipating power, as well as the redundancy and robustness to localized cell failure. We examine how the connectivity changes with the geometry of the conductive cell surface area, and in particular, the percentage of the cell half that is conductive and makes contact with neighboring cells. We find that the best connectivity exists when the conductive surface of the cell is approximately 80% of the hemisphere surface, addressing the tradeoff of maximizing contact with neighboring cells while minimizing shorts in the structure. In terms of robustness, the results show that, for the proposed circular and spherical cell design, the connectivity is a nearly linear function of the number of disconnects, indicating that there is not a catastrophic effect of isolated cell failures. In terms of structure size, the connectivity appears to plateau at around 60% for the planar structures and around 50% for the cubic structures of around 500 cells or greater with random cell orientation.

Empirical Characterization of Convergence Properties for Kernel-based Visual Servoing

Visual servoing typically involves separate feature tracking and control processes. Feature tracking remains an art, and is generally treated as independent of the underlying controller. Kernel-based visual servoing (KBVS) is a categorically different approach that eliminates explicit feature tracking. This chapter presents an experimental assessment of the convergence properties (domain of attraction and steady-state error) of the proposed approach. Using smooth weighting functions (the kernels) and Lyapunov theory, we analyze the controllers as they act on images acquired in controlled environments. We ascertain the domain of attraction by finding the largest positive invariant set of the Lyapunov function, inside which its time derivative is negative definite. Our experiments show that KBVS attains a maximum pixel error of one pixel and is commonly on the order of one tenth of a pixel.