Joris De Schutter

University of Leuven, Louvain, Flanders, Belgium

Are you Joris De Schutter?

Claim your profile

Publications (358)261.69 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces a framework for constraint-based force/position control of robots that exhibit large nonlinear structural compliance and that undergo large deformations. Controller synthesis follows hereto the principles of the Task Frame and instantaneous Task Specification using Constraints (iTaSC) formalisms. iTaSC is found particularly suitable due to its ability to express and combine control tasks in a natural way. Control tasks can be formulated as combinations of target positions, velocities, or forces expressed in an arbitrary number and type of coordinate frames. The proposed framework is applied to a mixed mechatronic system composed of a traditional rigid-link robot whose end-effector is a continuum (flexible) link. A selection of different position/force control tasks is prepared to demonstrate the validity and general nature of the proposed framework.
    IEEE Transactions on Robotics 10/2015; 31(5):1252-1260. DOI:10.1109/TRO.2015.2475975 · 2.43 Impact Factor
  • Source
    Maxim Vochten · Tinne De Laet · Joris De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents an overview and comparison of minimal and complete rigid body motion trajectory descriptors, usable in applications like motion recognition and programming by demonstration. Motion trajectory descriptors are able to deal with potentially unwanted variations acting on the motion trajectory such as changes in the execution time, the motion's starting position, or the viewpoint from which the motion is observed. A suitable rigid body motion trajectory descriptor retains only the trajectory information relevant to the application. This paper compares different trajectory descriptors for rigid body motion and validates their usefulness for dealing with motion variation in a motion recognition experiment. Furthermore, a new type of invariant trajectory descriptor is introduced based on the Frenet-Serret formulas.
    2015 IEEE International Conference on Robotics and Automation (ICRA), Seattle; 05/2015
  • Jonas Vantilt · Erwin Aertbelien · Friedl De Groote · Joris De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: An increasing number of robotic systems are using compliant actuators in which springs are placed in series with the actuator. The need for identification procedures tailored to these systems is consequently rising. When measurements of both the link side and the motor side of the spring are available the dynamic parameters can be identified independent of the spring model. The excitation of the identification procedure is optimized for the identification of the dynamical parameters to provide a rich data set by maximizing a reduced information matrix. Adopting this reduced matrix leads to a near-optimal excitation. A Fourier series is chosen as parametrization of the excitation, resulting in a periodic movement of the system under identification. This periodicity is exploited to develop an alternative weighting of the parameter estimation. This proposed weighting uses the motor torque variance at each time step of the trajectory, instead of one global torque variance. This identification procedure is demonstrated and validated on one leg of a lower limb exoskeleton. Only the dynamic model in the sagital plane (2D) is identified.
    2015 IEEE International Conference on Robotics and Automation, Seattle; 05/2015
  • Tjorven Delabie · Joris De Schutter · Bart Vandenbussche ·
    [Show abstract] [Hide abstract]
    ABSTRACT: In this paper, a star tracker attitude estimation procedure with increased robustness and efficiency, using the AIM algorithm, is presented and validated. The unique approach of the AIM algorithm allows one to introduce a reliable quality check that can be efficiently calculated. Unlike existing validation methods, this quality check not only detects that some of the data are unreliable but it also determines which star measurements are unreliable. These unreliable measurements can be removed from the data set, and a new attitude quaternion can be calculated without having to repeat the entire AIM algorithm. This greatly improves the robustness of the attitude estimation, while limiting the computational expense. Furthermore, the structure of AIM allows one to reuse previously calculated data when the change in attitude between subsequent measurements is small. This way, the efficiency of the entire attitude estimation cycle can be increased significantly. These enhancements are validated with simulated star tracker data. The results show that the improvements significantly improve the robustness and lower the computational cost of the star tracker attitude estimation. As a consequence, the overall performance of the attitude determination and control system greatly increases. The increased efficiency of the attitude estimation could also allow the use of star trackers in smaller satellite projects with smaller budgets. Copyright © 2014 by Tjorven Delabie. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.
    Journal of Guidance Control and Dynamics 04/2015; 38(11):1-14. DOI:10.2514/1.G000894 · 1.29 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Scaled generic musculoskeletal models are commonly used to drive dynamic simulations of motions. It is however, acknowledged that not accounting for variability in musculoskeletal geometry and musculotendon parameters may confound the simulation results, even when analysing control subjects. This study documents the three-dimensional anatomical variability of musculotendon origins and insertions of 33 lower limb muscles determined based on magnetic resonance imaging in six subjects. This anatomical variability was compared to the musculotendon point location in a generic musculoskeletal model. Furthermore, the sensitivity of muscle forces during gait, calculated using static optimization, to perturbations of the musculotendon point location was analyzed with a generic model. More specific, a probabilistic approach was used: for each analyzed musculotendon point, the three-dimensional location was re-sampled with a uniform Latin hypercube method within the anatomical variability and the static optimization problem was then re-solved for all perturbations. We found that musculotendon point locations in the generic model showed only variable correspondences with the anatomical variability. The anatomical variability of musculotendon point location did affect the calculated muscle forces: muscles most sensitive to perturbations within the anatomical variability are iliacus and psoas. Perturbation of the gluteus medius anterior, iliacus and psoas induces the largest concomitant changes in muscle forces of the unperturbed muscles. Therefore, when creating subject-specific musculoskeletal models, these attachment points should be defined accurately. In addition, the size of the anatomical variability of the musculotendon point location was not related to the sensitivity of the calculated muscle forces. Copyright © 2015. Published by Elsevier Ltd.
    Journal of Biomechanics 03/2015; 48(10). DOI:10.1016/j.jbiomech.2015.02.052 · 2.75 Impact Factor
  • Max Boegli · Tinne De Laet · Joris De Schutter · Jan Swevers ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents a model that closely approximates the generalized Maxwell-Slip (GMS) model, a multistate friction model known to describe all essential friction characteristics in presliding and sliding motion. In contrast to the GMS model, which consists of a switching structure to accommodate for its hybrid nature, the new model, referred to as the smoothed GMS (S-GMS) model, consists of an analytic set of differential equations. Such a model is well suited for gradient-based state and parameter estimation, as in the extended Kalman filter (EKF) or in moving horizon estimation. Similar to the GMS model, the S-GMS model is a multistate model that also describes all essential friction characteristics. Moreover, the S-GMS model description includes the single-state LuGre model and Elastoplastic model as special cases. This paper also discusses the implementation of the EKF estimator for the S-GMS friction model and compares its performance to the LuGre model in joint state and parameter estimation.
    IEEE/ASME Transactions on Mechatronics 10/2014; 19(5):1593-1602. DOI:10.1109/TMECH.2013.2288944 · 3.43 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: SUMMARYA new method for the estimation of subject-specific muscle–tendon parameters of the knee actuators based on dynamometry experiments is presented. The algorithm aims at estimating the tendon slack length and the optimal muscle fiber length by minimizing the difference between experimentally reproduced and model-based joint moments. The key innovative features are as follows: (i) the inclusion of a priori physiological knowledge to define a physiologically feasible set, the hot start for the optimization, and constraints for the optimization and (ii) the introduction of a new (affine) transformation of the muscle–tendon parameters, which greatly improves the numerical condition of the optimization.The influence of the initial guess and of measurement noise was studied in a simulation environment, and the performance was compared with that of the method presented earlier by Garner and Pandy for the upper limb. The tendon slack length was estimated for 97.5/63% (extensors/flexors) of all initial guesses within 2% of the ground truth. The optimal fiber length was estimated for 89/90% (extensors/flexors) of all initial guesses within 2% of the ground truth. When 10 Nm measurement noise was added, the mean value of the estimated tendon slack length deviated at most 1.9/1.6% (extensors/flexors) from the ground truth whereas the standard deviations were at most 5.1/3.9%. The mean value of the estimated optimal fiber length deviated at most 4.3/3.0% (extensors/flexors) from the ground truth whereas the standard deviations were at most 10.2/15.5%. In comparison, mean values resulting from the method of Garner and Pandy deviated up to 181% ( ± 123%) and 119% ( ± 30%) from the ground truth for, respectively, optimal fiber length and tendon slack length of rectus femoris.We concluded that the presented method had a low dependency on the initial guess and outperformed the method of Garner and Pandy in terms of accuracy by at least one order of magnitude when parameters were estimated from noisy data. The improvements open new perspectives for subject-specific modelling of muscles and tendons, which is beneficial for the accuracy of human motion simulations. Copyright © 2014 John Wiley & Sons, Ltd.
    10/2014; 30(10). DOI:10.1002/cnm.2639

  • ESMAC 2014, Rome; 09/2014
  • Erwin Aertbeliën · Joris De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Abstract— This paper presents a new framework for constraint-based task specification of robot controllers. A task specification language (eTaSL) is defined as well as a corre- sponding implementation of a controller (eTC). This new frame- work is based on feature variables and a new concept referred to as expression graphs. It avoids some of the common pitfalls in previous frameworks, and provides a flexible and composable way to define robot control tasks. An architecture for a robot controller is proposed, as well as an implementation that can execute tasks described in the new specification language. Typical usage patterns for the new framework are explained on an example consisting of a kinematically redundant, bi-manual task on a PR2 robot. A comparison with existing frameworks shows the advantages of the new approach.
    2014 IEEE/RSJ International Conference on Intelligent Robots and Systems, Chicago, IL, USA; 09/2014
  • Erwin Aertbeliën · Joris De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Abstract— For an intelligent dynamic motion interaction between a human and a lower-limb exoskeleton, it is necessary to predict the future evolution of the joint gait trajectories and to detect which phase of the gait pattern is currently active. A model of the gait trajectories and of the variations on these trajectories is learned from an example data set. A gait prediction module, based on a statistical latent variable model, is able to predict, in real-time, the future evolution of a joint trajectory, an estimate of the uncertainty on this prediction, the timing along the trajectory and the consistency of the measurements with the learned model. The proposed method is validated using a data set of 54 trials of children walking at three different velocities.
    5th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics (BIOROB 2014), São Paulo, Brazil; 08/2014
  • Source
    Maarten Afschrift · Friedl De Groote · Joris De Schutter · Ilse Jonkers ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Enabling persons with functional weaknesses to perform activities of daily living (ADL) is one of the main challenges for the aging society. Powered orthoses, or exoskeletons, have the potential to support ADL while promoting active participation of the user. For this purpose, assistive devices should be designed and controlled to deliver assistance as needed (AAN). This means that the level of assistance should bridge the capability gap, i.e. the gap between the capabilities of the subjects and the task requirements. However, currently the actuators of exoskeletons are mainly designed using inverse dynamics (ID) based calculations of joint moments. The goal of the present study is to calculate the capability gap for the lower limb during ADL when muscle weakness is present, which is needed for appropriate selection of actuators to be integrated in exoskeletons. Methods A musculoskeletal model (MM) is used to calculate the joint kinematics, joint kinetics and muscle forces of eight healthy subjects during ADL (gait, sit-to-stand, stand-to-sit, stair ascent, stair descent). Muscle weakness was imposed to the MM by a stepwise decrease in maximal isometric force imposed to all muscles. Muscle forces were calculated using static optimization. In order to compensate for muscle weakness, ideal moment actuators that represent the motors of an exoskeleton in the simulation were added to deliver AAN required to perform the task. Results The ID approach overestimates the required assistance since it relies solely on the demands of the task, whereas the AAN approach incorporates the capabilities of the subject. Furthermore, the ID approach delivers continuous support whereas the AAN approach targets the period where a capability gap occurs. The level of muscle weakness for which the external demands imposed by ADL can no longer be met by active muscle force production, is respectively 40%, 70%, 80% and 30%. Conclusions The present workflow allows estimating the AAN during ADL for different levels of muscle weakness, which can be used in the mechatronic design and control of powered exoskeletons. The AAN approach is a more physiological approach than the ID approach, since the MM accounts for the subject-specific capabilities of the user.
    BioMedical Engineering OnLine 08/2014; 13(1):111. DOI:10.1186/1475-925X-13-111 · 1.43 Impact Factor
  • Source
    Wannes Van Loock · Goele Pipeleers · Moritz Diehl · Joris De Schutter · Jan Swevers ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Path following deals with the problem of following a geometric path with no predefined timing information and constitutes an important step in solving the motion-planning problem. For differentially flat systems, it has been shown that the projection of the dynamics along the geometric path onto a linear single-input system leads to a small dimensional optimal control problem. Although the projection simplifies the problem to great extent, the resulting problem remains difficult to solve, in particular in the case of nonlinear system dynamics and time-optimal problems. This paper proposes a nonlinear change of variables, using a time transformation, to arrive at a fixed end-time optimal control problem. Numerical simulations on a robotic manipulator and a quadrotor reveal that the proposed problem formulation is solved efficiently without requiring an accurate initial guess.
    IEEE Transactions on Robotics 08/2014; 30(4):980-985. DOI:10.1109/TRO.2014.2305493 · 2.43 Impact Factor
  • Tjorven Delabie · Gert Raskin · Bart Vandenbussche · Joris De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: The quality of space telescope observations greatly depends on the pointing performance of the spacecraft. In this paper, recent advances in star tracker algorithms are discussed. This paper discusses efficient star tracker algorithms that improve the pointing performance of the satellite, resulting in observations of higher quality. Furthermore, the greatly reduced computational cost of these algorithms facilitates the inclusion of astronomical payload measurements in the attitude determination and control loop. When the payload is used as an additional star tracker, the pointing performance of the spacecraft increases drastically, which in its turn improves the quality of the scientific measurements. Simulations show that with these improvements, the absolute pointing error of the spacecraft can be reduced considerably.
    SPIE Astronomical Telescopes + Instrumentation; 07/2014
  • J. Verbeke · D. Hulens · H. Ramon · T. Goedeme · J. De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: We have conceived a novel compound multicopter (helicopter type utilizing multiple different size propellers for separate lift and attitude control) configuration specifically for flight through narrow corridors. Its design combines the contradictory requirements of limited width, high agility and long endurance while carrying a significant payload. This configuration can be scaled for both indoor and outdoor applications. The development is part of a doctoral research in which an autonomous unmanned rotary helicopter is designed, constructed and flight tested for inspecting fruit orchards and vineyards while flying in between the tree rows in outdoor conditions such as wind and gusts. The compound hexacopter configuration combines two large lift propellers, with a constant rotational velocity, with four small control propellers commanded by an autopilot. The autopilot is configured as a quadcopter commanding only the control propellers as only these change the attitude and overall thrust of the hexacopter. The benefit of using large lift propellers is their lower disk loading (thrust divided by disk area) which results in a higher Figure of Merit and lower power consumption compared to the smaller control propellers, while the latter are better suited for outdoor (windy) conditions due to their fast reaction time in spooling up and down. Compared to a standard quadcopter with the same width, payload and battery capacity, the endurance of the compound hexacopter is potentially up to 60% higher. As a concept validator, a small-scale prototype has been designed, constructed and successfully flight tested.
    2014 International Conference on Unmanned Aircraft Systems (ICUAS); 05/2014
  • Gianni Borghesan · Erwin Aertbeliën · Joris De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: bstract—This work aims to extend the application field ofthe constraint-based control framework called iTaSC (instanta-neous task specification using constraints) toward manipulationtasks. iTaSC offers two advantages with respect to other meth-ods: the ability to specify tasks in different spaces (and not onlyin Cartesian coordinates as for the Task Frame Formalism),and the treatment of geometric uncertainties. These propertiesmay be very useful within a manipulation context, where tasksare executed by robots with many degrees of freedom, whichcalls for some degree of abstraction; by choosing a suitableset of coordinates, it is possible to reduce the complexity andthe number of constraints that fully describe such tasks; inaddition, controlling only the subspace that is needed to fulfila task allows us to use the remaining degrees of freedom ofthe robot system to achieve secondary objectives. This paperdiscusses the instruments and techniques that can be employedin manipulation scenarios; in particular it focuses on aspectslike the specification of a grasp and control of the stance of therobotic arm.iTaSC offers the possibility of specifying a grasp. While thisapproach allows for very fine control of a grasping task, inmost cases a less fine-grain specification suffices to guarantee asuccessful execution of the grasping action. To this end synergy- based grasp specification is formulated within iTaSC.We also show how to take into account secondary objectives for the arm stance. In particular we consider, as an example,the manipulability index along a given direction. Such indexesare maximised by exploring the null space of the other tasks.The proposed approach is demonstrated by means of simu-lations, where a robotic hand grasps a cylindrical object.
    Tthe International Conference on Robotics and Automation (ICRA2014), Hong Kong, China; 05/2014
  • Source
    B. Theys · G. Dimitriadis · T. Andrianne · P. Hendrick · J. De Schutter ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper presents experimental results of the full 3-axis force vector and 3-axis moment vector acting on a propeller, commonly used for a Vertical Take Off and Landing Micro Aerial Vehicle (VTOL MAV). Measurements were carried out in a wind tunnel using a high resolution 6-axis force/moment sensor embedded in a customized test rig at several wind speeds, propeller rotational speeds and angles of the propeller shaft with respect to the air stream. Results show strong moments acting on the propeller in forward flight and unstable conditions in descending flight. Power calculations reveal a decrease in power consumption during slow forward flight and how motor efficiency can be maximized.
    2014 International Conference on Unmanned Aircraft Systems (ICUAS); 05/2014
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background Spasticity is an important complication after stroke, especially in the anti-gravity muscles, i.e. lower limb extensors. However the contribution of hyperexcitable muscle spindle reflex loops to gait impairments after stroke is often disputed. In this study a neuro-musculoskeletal model was developed to investigate the contribution of an increased length and velocity feedback and altered reflex modulation patterns to hemiparetic gait deficits. Methods A musculoskeletal model was extended with a muscle spindle model providing real-time length and velocity feedback of gastrocnemius, soleus, vasti and rectus femoris during a forward dynamic simulation (neural control model). By using a healthy subject’s base muscle excitations, in combination with increased feedback gains and altered reflex modulation patterns, the effect on kinematics was simulated. A foot-ground contact model was added to account for the interaction effect between the changed kinematics and the ground. The qualitative effect i.e. the directional effect and the specific gait phases where the effect is present, on the joint kinematics was then compared with hemiparetic gait deviations reported in the literature. Results Our results show that increased feedback in combination with altered reflex modulation patterns of soleus, vasti and rectus femoris muscle can contribute to excessive ankle plantarflexion/inadequate dorsiflexion, knee hyperextension/inadequate flexion and increased hip extension/inadequate flexion during dedicated gait cycle phases. Increased feedback of gastrocnemius can also contribute to excessive plantarflexion/inadequate dorsiflexion, however in combination with excessive knee and hip flexion. Increased length/velocity feedback can therefore contribute to two types of gait deviations, which are both in accordance with previously reported gait deviations in hemiparetic patients. Furthermore altered modulation patterns, in particular the reduced suppression of the muscle spindle feedback during swing, can contribute largely to an increased plantarflexion and knee extension during the swing phase and consequently to hampered toe clearance. Conclusions Our results support the idea that hyperexcitability of length and velocity feedback pathways, especially in combination with altered reflex modulation patterns, can contribute to deviations in hemiparetic gait. Surprisingly, our results showed only subtle temporal differences between length and velocity feedback. Therefore, we cannot attribute the effects seen in kinematics to one specific type of feedback.
    Journal of NeuroEngineering and Rehabilitation 04/2014; 11(1):78. DOI:10.1186/1743-0003-11-78 · 2.74 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Human body detection and pose estimation is useful for a wide variety of applications and environments. Therefore a human body detection and pose estimation system must be adaptable and customizable. This paper presents such a system that extracts skeletons from RGB-D sensor data. The system adapts on-line to difficult unstructured scenes taken from a moving camera (since it does not require background subtraction) and benefits from using both color and depth data. It is customizable by virtue of requiring less training data, having a clearly described training method, and a customizable human kinematic model. Results show successful application to data from a moving camera in cluttered indoor environments. This system is open-source, encouraging reuse, comparison, and future research.
    Journal of Visual Communication and Image Representation 01/2014; 25(1):39–52. DOI:10.1016/j.jvcir.2013.03.011 · 1.22 Impact Factor
  • Source
    Tinne De Laet · Joris De Schutter · Herman Bruyninckx ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper proposes and experimentally validates a Bayesian network model of a range finder adapted to dynamic environments. All modeling assumptions are rigorously explained, and all model parameters have a physical interpretation. This approach results in a transparent and intuitive model. With respect to the state of the art beam model this paper: (i) proposes a different functional form for the probability of range measurements caused by unmodeled objects, (ii) intuitively explains the discontinuity encountered in te state of the art beam model, and (iii) reduces the number of model parameters, while maintaining the same representational power for experimental data. The proposed beam model is called RBBM, short for Rigorously Bayesian Beam Model. A maximum likelihood and a variational Bayesian estimator (both based on expectation-maximization) are proposed to learn the model parameters. Furthermore, the RBBM is extended to a full scan model in two steps: first, to a full scan model for static environments and next, to a full scan model for general, dynamic environments. The full scan model accounts for the dependency between beams and adapts to the local sample density when using a particle filter. In contrast to Gaussian-based state of the art models, the proposed full scan model uses a sample-based approximation. This sample-based approximation enables handling dynamic environments and capturing multi-modality, which occurs even in simple static environments.
    Journal of Artificial Intelligence Research 01/2014; 33. DOI:10.1613/jair.2540 · 1.26 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Time-optimal path following considers the problem of moving along a predetermined geometric path in minimum time. In the case of a robotic manipulator with simplified constraints, a convex reformulation of this optimal control problem has been derived previously. However, many applications in robotics feature constraints such as velocity-dependent torque constraints or torque rate constraints that destroy the convexity. The present paper proposes an efficient sequential convex programming (SCP) approach to solve the corresponding nonconvex optimal control problems by writing the nonconvex constraints as a difference of convex (DC) functions, resulting in convex-concave constraints. We consider seven practical applications that fit into the proposed framework even when mutually combined, illustrating the flexibility and practicality of the proposed framework. Furthermore, numerical simulations for some typical applications illustrate the fast convergence of the proposed method in only a few SCP iterations, confirming the efficiency of the proposed framework.
    IEEE Transactions on Robotics 12/2013; 29(6):1485-1495. DOI:10.1109/TRO.2013.2277565 · 2.43 Impact Factor

Publication Stats

4k Citations
261.69 Total Impact Points


  • 1-2015
    • University of Leuven
      • Department of Mechanical Engineering
      Louvain, Flanders, Belgium
  • 2014
    • University of Kansas
      Lawrence, Kansas, United States
  • 2001
    • University of North Carolina at Charlotte
      • Department of Computer Science
      Charlotte, North Carolina, United States
    • De La Salle University
      • Department of Mechanical Engineering
      Manila, NCR, Philippines
    • Center for Scientific Review
      베서스다, Maryland, United States
  • 1999
    • La Salle University
      Philadelphia, Pennsylvania, United States
  • 1996-1998
    • Belgian Nuclear Research Centre
      Moll, Flemish, Belgium
    • University of New South Wales
      • School of Mechanical and Manufacturing Engineering
      Kensington, New South Wales, Australia