The use of an HTC Vive; a virtual reality (VR) system and its innovative tracking technology is explored in order to create an approximate one-to-one mapping to the virtual representation of a robot cell. The mapping is found by performing hand-eye calibration, establishing a spatial relationship between the inertial frames of the robot cell and the tracking system. One of the main contributions of this article is the development of an open-source Robotic Operating System (ROS) package for VR devices such as the Vive. The package includes automated calibration procedures such that the devices gives a centimetric measurement error in the robot cell. The calibrated system has problems that are related to specific issues of the tracking technology. This article outlines these issues, their cause, and potential fixes in a concise manner. A simple assembly scenario is presented, where the outline of objects in the robot cell are defined by registering points with a Vive tracker. The potential use cases of the calibrated system are limited by its accuracy, and depends on the required tolerances.

A system architecture is presented to generate sensor-controlled robot tasks from knowledge encoded in a CAD model. This architecture consists of an application layer where the user annotates assembly tasks in the CAD software. A process layer infers the specific robot skills and parameters from the CAD model and annotated data. A control layer executes the complex, force-controlled tasks. A proof-of-concept implementation is made, consisting of an application layer implemented in FreeCAD and a process layer that focuses on using fuzzy inference to generate appropriate skill-dependent process parameters from the geometric CAD information and annotations in the CAD model. In the control layer, a constraintbased control framework is used to robustly execute the assembly tasks. The system is validated on a challenging assembly task involving the assembly of screw compressor parts.

Fast, accurate evaluation of the dynamics parameters is a key ingredient for accurate control, estimation, and simulation of robots. As these are time-consuming to compute by hand, a software library for generating the rigid body dynamics symbolically can be of great use for robotics researchers. In this paper, we propose a library to efficiently compute and evaluate robot dynamics and its derivatives. Based on a URDF description of the robot's kinematics, three major rigid body dynamics algorithms are used to retrieve the dynamics symbolically in the CasADi framework. To validate the numerical accuracy, the numerical evaluation of the solutions are compared against three other well-established rigid body dynamics libraries, namely RBDL, KDL, and PyBullet. We conduct a timing comparison between the libraries, and we show that the evaluation times of the symbolic expressions are at most one order of magnitude higher than the evaluation times of the numerical libraries. Last, it is shown that the evaluation times of the dynamics derivatives remain of the same order as the evaluation times of the dynamics expressions.

Abstract-In this article the model predictive path following controller and the model predictive trajectory tracking controller are compared for a robotic manipulator. We consider both the Runge-Kutta and collocation based discretization. We show how path-following can stop at obstructions in a way trajectory tracking cannot. We give simulations for a twolink manipulator, and discuss the real-time viability of our implementations.

Visual Odometry (VO) is increasingly a useful tool for robotic navigation in a variety of applications, including weed removal for agricultural robotics. The methods of evaluating VO are often computationally expensive and can cause the VO measurements to be significantly delayed with respect to a compass, wheel odometry, and GPS measurements. In this paper we present a Bayesian formulation of fusing delayed displacement measurements. We implement solutions to this problem based on the unscented Kalman filter (UKF), leading to what we term an unscented multi-point smoother. The proposed methods are tested in simulations of an agricultural robot. The simulations show improvements in the localization RMS error when including the VO measurements with a variety of latencies.