T. De Laet

Kalman smoothing improves the estimation of joint kinematics and kinetics in marker-based human gait analysis.

We developed a Kalman smoothing algorithm to improve estimates of joint kinematics from measured marker trajectories during motion analysis. Kalman smoothing estimates are based on complete marker trajectories. This is an improvement over other techniques, such as the global optimisation method (GOM), Kalman filtering, and local marker estimation (LME), where the estimate at each time instant is only based on part of the marker trajectories. We applied GOM, Kalman filtering, LME, and Kalman smoothing to marker trajectories from both simulated and experimental gait motion, to estimate the joint kinematics of a ten segment biomechanical model, with 21 degrees of freedom. Three simulated marker trajectories were studied: without errors, with instrumental errors, and with soft tissue artefacts (STA). Two modelling errors were studied: increased thigh length and hip centre dislocation. We calculated estimation errors from the known joint kinematics in the simulation study. Compared with other techniques, Kalman smoothing reduced the estimation errors for the joint positions, by more than 50% for the simulated marker trajectories without errors and with instrumental errors. Compared with GOM, Kalman smoothing reduced the estimation errors for the joint moments by more than 35%. Compared with Kalman filtering and LME, Kalman smoothing reduced the estimation errors for the joint accelerations by at least 50%. Our simulation results show that the use of Kalman smoothing substantially improves the estimates of joint kinematics and kinetics compared with previously proposed techniques (GOM, Kalman filtering, and LME) for both simulated, with and without modelling errors, and experimentally measured gait motion.

iTASC: a tool for multi-sensor integration in robot manipulation

iTASC (acronym for dasiainstantaneous task specification and controlpsila) by J. De Schutter (2007) is a systematic constraint-based approach to specify complex tasks of general sensor-based robot systems. iTASC integrates both instantaneous task specification and estimation of geometric uncertainty in a unified framework. Automatic derivation of controller and estimator equations follows from a geometric task model that is obtained using a systematic task modeling procedure. The approach applies to a large variety of robot systems (mobile robots, multiple robot systems, dynamic human-robot interaction, etc.), various sensor systems, and different robot tasks. Using an example task, this paper shows that iTASC is a powerful tool for multi-sensor integration in robot manipulation. The example task includes multiple sensors: encoders, a force sensor, cameras, a laser distance sensor and a laser scanner. The paper details the systematic modeling procedure for the example task and elaborates on the task specific choice of two types of task coordinates: feature coordinates, defined with respect to object and feature frames, which facilitate the task specification, and uncertainty coordinates to model geometric uncertainty. Experimental results for the example task are presented.

Geometric Relations Between Rigid Bodies (Part 1): Semantics for Standardization

This tutorial explicitly states the semantics of all coordinate-invariant properties and operations, and, more importantly, all the choices that are made in coordinate representations of these geometric relations. This results in a set of concrete suggestions for standardizing terminology and notation, allowing programmers to write fully unambiguous software interfaces, including automatic checks for semantic correctness of all geometric operations on rigid-body coordinate representations. A concrete proposal for community-driven standardization via the Robot Engineering Task Force [4] is accepted as a Robotics Request for Comment.

Shape-Based Online Multitarget Tracking and Detection for Targets Causing Multiple Measurements: Variational Bayesian Clustering and Lossless Data Association

This paper proposes a novel online two-level multitarget tracking and detection (MTTD) algorithm. The algorithm focuses on multitarget detection and tracking for the case of multiple measurements per target and for an unknown and varying number of targets. Information is continuously exchanged in both directions between the two levels. Using the high level target position and shape information, the low level clusters the measurements. Furthermore, the low level features automatic relevance detection (ARD), as it automatically determines the optimal number of clusters from the measurements taking into account the expected target shapes. The high levels data association allows for a varying number of targets. A joint probabilistic data association algorithm looks for associations between clusters of measurements and targets. These associations are used to update the target trackers and the target shapes with the individual measurements. No information is lost in the two-level approach since the measurement information is not summarized into features. The target trackers are based on an underlying motion model, while the high level is supplemented with a filter estimating the number of targets. The algorithm is verified using both simulations and experiments using two sensor modalities, video and laser scanner, for detection and tracking of people and ants.

Rigorously Bayesian range finder sensor model for dynamic environments

This paper proposes and experimentally validates a Bayesian network model of a range finder adapted to dynamic environments. The modeling rigorously explains all model assumptions and parameters, improving the physical interpretation of all parameters and the intuition behind the model choices. With respect to the state of the art model [1], this paper proposes: (i) a different functional form for the probability of range measurements caused by unexpected objects, (ii) an intuitive explanation for the discontinuity encountered in the cited paper, and (iii) a reduction in the number of model parameters, while maintaining the same representational power for experimentally obtained data. The proposed beam model is called RBBM, short for rigorously Bayesian beam model. A maximum-likelihood estimation and a variational Bayesian estimation algorithm (both based on expectation-maximization) are proposed to learn the model parameters.

Geometric Relations Between Rigid Bodies (Part 2): From Semantics to Software

Rigid bodies are essential primitives in the modeling of robotic devices, tasks, and perception. Basic geometric relations between rigid bodies include relative position, orientation, pose, linear velocity, angular velocity, twist, force, torque, and wrench. In Part 1 of this tutorial [3], we explicitly stated the semantics of all coordinate-invariant properties and operations, and, more importantly, all the choices that are made in coordinate representations of these geometric relations. This resulted in a set of concrete suggestions for standardizing terminology and notation.

An application of constraint-based task specification and estimation for sensor-based robot systems

This paper shows the application of a systematic approach for constraint-based task specification for sensor-based robot systems to a laser tracing example. This approach integrates both task specification and estimation of geometric uncertainty in a unified framework. The framework consists of an application independent control and estimation scheme. An automatic derivation of controller and estimator equations is achieved, based on a geometric task model that is obtained using a systematic task modeling procedure. The paper details the systematic modeling procedure for the laser tracing task and elaborates on the task specific choice of two types of task coordinates: feature coordinates, defined with respect to object and feature frames, which facilitate the task specification, and uncertainty coordinates to model geometric uncertainty. Furthermore, the control and estimation scheme for this specific task is studied. Simulation and real world experimental results are presented for the laser tracing example.

Comparison between bridging students and traditional first-year students in engineering technology

ABSTRACT To stimulate a flexible lifelong learning system students can enter university via lateral entry. Unlike traditional first-year students, lateral entrance students are not well-studied. Therefore this study focuses on comparing first-year students with a specific group of lateral entrants, namely bridging students at the Faculty of Engineering Technology, KU Leuven. Using Astin’s Input-Environment-Outcome model resulted in (1) Input variables, namely prior education and initial learning and study strategies, (2) Environmental influence, measured with a questionnaire focussing on perceived transition to university, and (3) Outcome variables, namely dropout and academic achievement. Analyses resulted in similarities for the outcome variables, but differences in terms of secondary education. Regarding the input (LASSI) and environmental questionnaires, for only two of the 13 scales a moderate effect was found (perceived preparedness and test strategies). Consequently, research findings of first-year engineering students can be compared, taking into account their specific differences, to the context of bridging students.

Application of a Generic Constraint-Based Programming Approach to an Industrially Relevant Robot Task with Geometric Uncertainties

This paper shows the application of a generic constraint-based task specification approach for sensor-based robot systems to a laser tracing example. Key properties of the used approach are (i) its ability to specify complex robot tasks by introducing auxiliary task-oriented feature coordinates, defined with respect to user-defined object and feature frames, (ii) its support for both underconstrained and overconstrained robot tasks, and (iii) its ability to integrate sensor measurements in a unified way, using auxiliary uncertainty coordinates, to estimate geometric uncertainties in the robot system or its environment. Simulation and real world experimental results are presented.