Movement templates for learning of hitting and batting
The DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index.
Abstract
Hitting and batting tasks, such as tennis forehands, ping-pong strokes, or baseball batting, depend on predictions where the ball can be intercepted and how it can properly be returned to the opponent. These predictions get more accurate over time, hence the behaviors need to be continuously modified. As a result, movement templates with a learned global shape need to be adapted during the execution so that the racket reaches a target position and velocity that will return the ball over to the other side of the net or court. It requires altering learned movements to hit a varying target with the necessary velocity at a specific instant in time. Such a task cannot be incorporated straightforwardly in most movement representations suitable for learning. For example, the standard formulation of the dynamical system based motor primitives (introduced by Ijspeert et al. [1]) does not satisfy this property despite their flexibility which has allowed learning tasks ranging from locomotion to kendama. In order to fulfill this requirement, we reformulate the Ijspeert framework to incorporate the possibility of specifying a desired hitting point and a desired hitting velocity while maintaining all advantages of the original formulation. We show that the proposed movement template formulation works well in two scenarios, i.e., for hitting a ball on a string with a table tennis racket at a specified velocity and for returning balls launched by a ball gun successfully over the net using forehand movements. All experiments were carried out on a Barrett WAM using a four camera vision system.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
A New Approach to Linear Filtering and Prediction Problems
Kalman, R. E. (1960) · Journal of Basic Engineering · doi:10.1115/1.3662552 · Cited by 31,751
Movement imitation with nonlinear dynamical systems in humanoid robots
Ijspeert, A. J., Nakanishi, J., & Schaal, S. (2003) · Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292) · doi:10.1109/robot.2002.1014739 · Cited by 840 · cited 6×
“While the original formulation in [1], [2] for discrete dynamical systems motor primitives used a second-order system to represent the phase z of the movement, this formulation has proven to be unnecessarily complicated in practice.”
BackgroundMethodInfluentialMotor Learning and Performance
Schmidt, R. A., & Wrisberg, C. A. (1991) · Cited by 664
Timing an attacking forehand drive in table tennis.
Bootsma, R. J., & Van Wieringen, P. C. (1990) · Journal of Experimental Psychology: Human Perception and Performance · doi:10.1037//0096-1523.16.1.21 · Cited by 546
Policy Gradient Methods for Robotics
Peters, J., & Schaal, S. (2006) · 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems · doi:10.1109/iros.2006.282564 · Cited by 498 · cited 2×
“Although some of the presented examples, e.g., the tennis swing [1] or the T-ball batting [8], are striking movements, these standard motor primitives cannot properly encode a hitting movement.”
BackgroundSelf-citationLearning from demonstration and adaptation of biped locomotion
Nakanishi, J., Morimoto, J., Endo, G., Cheng, G., Schaal, S., & Kawato, M. (2004) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2004.03.003 · Cited by 399 · cited 1×
“Resulting successes have shown that it is possible to rapidly learn motor primitives for complex behaviors such as tennis swings [1] with only a final target, constrained reaching [4], drumming [5], biped locomotion [2], [6] and even in tasks with potential industrial application [7].”
BackgroundSchaal, S., Peters, J., Nakanishi, J., & Ijspeert, A. (2005) · Springer Tracts in Advanced Robotics · doi:10.1007/11008941_60 · Cited by 343
Self-citationDynamics systems vs. optimal control — a unifying view
Schaal, S., Mohajerian, P., & Ijspeert, A. (2007) · Progress in Brain Research · doi:10.1016/s0079-6123(06)65027-9 · Cited by 260 · cited 7×
“Since then, it has been simplified and in [3] it was shown that a single first order system suffices ˙ z = − τα z z. (1) This canonical system has the time constant τ = 1 /T where T is the duration of the motor primitive, a parameter α z which is chosen such that z ≈ 0 at T to ensure that the influence…”
BackgroundMethodInfluentialMovement reproduction and obstacle avoidance with dynamic movement primitives and potential fields
Park, D.-H., Hoffmann, H., Pastor, P., & Schaal, S. (2008) · Humanoids 2008 - 8th IEEE-RAS International Conference on Humanoid Robots · doi:10.1109/ichr.2008.4755937 · Cited by 254 · cited 1×
“The transformation function f ( z ) alters the output of the first system, in Eq.”
BackgroundA Robot Ping-Pong Player: Experiments in Real-Time Intelligent Control
Anderson, R. L. (1988) · Medical Entomology and Zoology · Cited by 162
Research works printed but not linked: 3
- ramanantsoa, 1994, Towards a stroke construction model, Int J Table Tennis Sci, 2, 97
- tyldesley, 1975, Operational timing, Journal of Human Movement Studies, 1, 172
- schaal, 2003, Learning Motor Primitives, Proceedings of the International Symposium on Robotics Research
Cited by
Items 101–154 of 154
Experiments with Motor Primitives in Table Tennis
Peters, J., Mülling, K., & Kober, J. (2013) · Springer tracts in advanced robotics · doi:10.1007/978-3-642-28572-1_24 · Cited by 5
Self-citationLiendo, F., Hernández, C., & Galez, C. (2025) · Journal of Mechanisms and Robotics · doi:10.1115/1.4068555 · Cited by 4
Optimal control with state and command limits for a simulated ball batting task
Schüthe, D., & Frese, U. (2015) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2015.7353939 · Cited by 4
“After a process overview in the previous section we describe the robot model our controller and the simulation is based on.”
MethodAn open-architecture integration solution for the research and development of smart robots
Do, T. T., Dinh, V. P., Nguyen, Q. H., Chu, D. H., & Ministry of Science and Technology (2022) · Ministry of Science and Technology Vietnam · doi:10.31276/vjste.64(4).36-44 · Cited by 4
Optimal trajectory generation for generalization of discrete movements with boundary conditions
Herzog, S., Wörgötter, F., & Kulvičius, T. (2016) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2016.7759486 · Cited by 4
“For comparison with DMPs we used a version of DMPs for generation of hitting and batting movements [8] since it allows generalization to both position and velocity boundary conditions.”
MethodLearning Motion Trajectories from Phase Space Analysis of the Demonstration
Gesel, P., Begum, M., & LaRoche, D. P. (2019) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra.2019.8794381 · Cited by 4
“Modifications of the original DMP formulations have been proposed to model such behavior [12], [11].”
BackgroundOptimizing robot striking movement primitives with Iterative Learning Control
Koç, O., Maeda, G. J., Neumann, G. M., & Peters, J. (2015) · IEEE-RAS International Conference on Humanoid Robots (Humanoids) · doi:10.1109/humanoids.2015.7363535 · Cited by 3
“Discrete DMPs have been extended in [16], [22] for striking movements, where the goal state is also evolved in time to allow for high velocities around settling time.”
BackgroundSelf-citationTesting probabilistic equivalence through Reinforcement Learning
Desharnais, J., Laviolette, F., & Zhioua, S. (2013) · Information and Computation · doi:10.1016/j.ic.2013.02.002 · Cited by 3
Motion learning from observation using Affinity Propagation clustering
Chang, G., & Kulić, D. (2013) · IEEE RO-MAN · doi:10.1109/roman.2013.6628424 · Cited by 4
“Motion learning consists of motion segmentation, motion modeling and motion clustering.”
Task Level Optimal Control of a Simulated Ball Batting Robot
Schüthe, D., & Frese, U. (2014) · International Conference on Informatics in Control, Automation and Robotics (ICINCO) · doi:10.5220/0005026100450056 · Cited by 3
“A lot of work concerning ball hitting tasks focuses on learning strategies (Kober et al., 2010; Muelling et al., 2013), where basic movements are learned by teach in and later adapted to the situation, or are based on trajectory planning in conjunction with a joint controller to follow them as described in (Senoo et a…”
BackgroundMotion Planning of a Dual Manipulator System for Table Tennis
Zhang, G., Wang, C., Li, B., & Zheng, H. (2012) · Advances in intelligent systems and computing · doi:10.1007/978-3-642-33932-5_32 · Cited by 3
“Kober et al. studied the hitting and batting tasks using the seven DOF Bar-rett WAM robot [7].”
MethodLearning Prioritized Control of Motor Primitives
Kober, J., & Peters, J. (2013) · Springer tracts in advanced robotics · doi:10.1007/978-3-319-03194-1_6 · Cited by 3
Self-citationFlexDMP – Extending Dynamic Movement Primitives towards Flexible Joint Robots
Wahrburg, A., Guida, S., Enayati, N., Zanchettin, A. M., & Rocco, P. (2021) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48506.2021.9560843 · Cited by 3
“Challenges are foreseen with respect to non-zero terminal velocities [4] and explicit speed constraints [9].”
BackgroundLiu, H., Li, Z., Wang, B., Zhou, Y., & Zhang, Q. (2013) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio.2013.6739577 · Cited by 3
“Mülling and Kober developed another robot table tennis system with an anthropomorphic 7-DOF arm as it manipulator[8], [9].”
BackgroundZJUNlict: RoboCup 2013 Small Size League Champion
Zhao, Y., Xiong, R., Tong, H., Li, C., & Fang, L. (2014) · Lecture notes in computer science · doi:10.1007/978-3-662-44468-9_9 · Cited by 3
“Moreover, we can change the target velocity of the movement while maintaining the overall duration and shape by developing several extensions and modifications to this approach as in [13].”
MethodExtending Dynamic Movement Primitives towards High-Performance Robot Motion
Wahrburg, A., Guida, S., Enayati, N., Zanchettin, A. M., & Rocco, P. (2020) · IEEE International Workshop on Advanced Motion Control (AMC) · doi:10.1109/amc44022.2020.9244421 · Cited by 3
“Numerous extensions of the DMP framework have been proposed, including non-zero terminal velocity [5], incorporating via-points and speed constraints [6], application of DMPs to orientation in Cartesian space [7], [8], and encoding of probabilistic movement primitives (ProMPs) [9].”
MethodLearning of Multi-Dimensional, Multi-Modal Features for Robotic Grasping
Detry, R. J. (2010) · ORBi (University of Liège) · Cited by 3
MBB: Model-Based Baseline for Efficient Reinforcement Learning.Preprint
Lyu, X., Li, S., Siriya, S., Pu, Y., & Chen, M. (2020) · arXiv (Cornell University) · Cited by 3
Motion-Primitive based Deep Reinforcement Learning for High Speed Aerospace Vehicle Missions
Levin, W. C., Nolan, S. M., Raz, A. K., Ezra, K., Parish, J. J., & Williams, K. (2023) · AIAA SciTech Forum · doi:10.2514/6.2023-2667 · Cited by 2
Rothomphiwat, K., Jaroonsorn, P., Kriengkomol, P., & Manoonpong, P. (2024) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros58592.2024.10801311 · Cited by 2
“The algo-rithm brings together the best aspects of previously modified dynamic movement primitives from [18], [19], and [16].”
Learning to Play Mini-Golf from Human Demonstration using Autonomous Dynamical Systems
Khansari-Zadeh, S. M., & Billard, A. (2011) · Infoscience (Ecole Polytechnique Fédérale de Lausanne) · Cited by 2
Spatio-temporal perception nets
Pongratz, M., Velik, R., & Machajdik, J. (2011) · IEEE AFRICON · doi:10.1109/afrcon.2011.6072061 · Cited by 2
“Current systems map physical models to their sensory data and apply mathematical or statistical models in order to act on this information [7], [8], [9].”
BackgroundReinforcement Learning of Phase Oscillators for Fast Adaptation to Moving Targets
Maeda, G. J., Koç, O., & Morimoto, J. (2018) · MPG.PuRe (Max Planck Society) · Cited by 2
Learning Visuomotor Policies with Deep Movement Primitives
Theofanidis, M., Bozcuoğlu, A. K., & Kyrarini, M. (2021) · International Conference on PErvasive Technologies Related to Assistive Environments (PETRA) · doi:10.1145/3453892.3453899 · Cited by 2
“In short, a DMP behaves like a nonlinear spring damper system [10] that pulls a degree of freedom to a goal position while following a learned trajectory.”
BackgroundLearning Sequential Skills for Robot Manipulation Tasks
Manschitz, S. (2018) · Technischen Universität Darmstadt · doi:10.26083/tuprints-00007185 · Cited by 3
Utilizing Human Feedback for Primitive Optimization in Wheelchair TennisPreprint
Krishna, A., Zaidi, Z., Chen, L., Paleja, R., Seraj, E., & Gombolay, M. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2212.14403 · Cited by 2
“Prior work has shown how movement primitives obtained from kinesthetic teaching can be used to teach robots to hit table tennis strokes [9, 10, 11, 12].”
BackgroundControlling the locomotion of quadruped robots with learning methods
Aractingi, M. (2023) · theses.fr (ABES) · doi:10.70675/d46a9cd8zd72cz4b1bzaa8ez71e0c6b58928 · Cited by 2
Task-oriented Sequential Pose Motion Primitives
Fu, J. Y., & Wang, N. (2023) · International Annual Conference on Complex Systems and Intelligent Science (CSIS-IAC) · doi:10.1109/csis-iac60628.2023.10364224 · Cited by 1
“In the study of batting motion [6], it is necessary to con-trol the speed of the bat.”
BackgroundWang, Y., Sun, Z., Qu, C., Luo, Y., Zhang, H., Dong, K., Liu, Y., Zhang, Q., Cai, B., Cheng, E., et al. (2025) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2025.3607764 · Cited by 1
“Kober et al. [20] reformulated the canonical dynamic movement primitives (DMPs) framework by Ijspeert et al. [21] and Schaal et al. [22], [23] to incorporate the ability to generate a desired hitting point and velocity.”
Distance-based Kernels for Dynamical Movement Primitives
Escudero-Rodrigo, D., & René, A. (2015) · Frontiers in artificial intelligence and applications · doi:10.3233/978-1-61499-578-4-133 · Cited by 1
“) seems a good choice, because they are compact parameterizations of the robot control policy and a well-established approach for representing modular and re-usable robot movement generators [2][3].”
BackgroundSolving Complex Tasks Hierarchically from Demonstrations
Zheng, W., Wu, B., & Lin, H. (2018) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc.2018.8431734 · Cited by 1
“In LfD, movement primitives are widely used for representing and learning basic movements in robotics [2][3][4].”
BackgroundSRC: RoboCup 2017 Small Size League Champion
Wei, R., Ma, W., Yu, Z., Huang, W. W., & Shan, S. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-00308-1_34 · Cited by 1
Decision Making Based on Physical and Neural Network Models for Precision Ball-Batting Robots
Hsiao, T., & Wu, S.-C. (2021) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc50511.2021.9482718 · Cited by 1
Research on Operation Skill Transfer and Control Oriented to Complex Force Interaction Tasks
Jie, Z., Rui, W., He, Z., Yanhe, Z., & Xizhe, Z. (2022) · Journal of Mechanical Engineering · doi:10.3901/jme.2022.18.116 · Cited by 2
Decision Making of Ball-Batting Robots Based on Deep Reinforcement Learning
Hsiao, T., & Kao, H.-C. (2023) · Proceedings of the ... American Control Conference/Proceedings of the American Control Conference · doi:10.23919/acc55779.2023.10156522 · Cited by 1
“For example, imitation learning was applied to learn the table-tennis strokes in [16],”
MethodDetermining predictable strike points on tossed objects: A 2D physics simulation approach
Giladi, C., & Golan, Y. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107268 · Cited by 1
Optimal Control-Based Baseline for Guided Exploration in Policy Gradient MethodsPreprint
Lyu, X., Li, S., Siriya, S., Pu, Y., & Chen, M. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.02073 · Cited by 1
“Such demonstrations can come from humans or dynamical systems, and the initialized policy can be created through supervised learning or explicit programming [18], [19].”
BackgroundDexterous robotic manipulation: Batting and cutting
Mu, X. (2021) · doi:10.31274/td-20240329-239 · Cited by 1
Motivations intrinsèques et contraintes maturationnelles pour l'apprentissage sensorimoteur
Baranès, A. (2011) · theses.fr (ABES) · doi:10.70675/64c82975zca03z4f0ezaf60zbb6b7f1b5cb8 · Cited by 5
Learning to Control Planar Hitting Motions of a Robotic Arm in a Mini-Golf-like Task
Kronander, K. (2010) · Infoscience (Ecole Polytechnique Fédérale de Lausanne)
Yin, X., & Chen, Q. (2015) · Conference proceedings/Conference proceedings - IEEE International Conference on Systems, Man, and Cybernetics · doi:10.1109/smc.2015.160
Ball Balancing Along a Trajectory
Gast, J., Darmstadt, T., Wild, B., & Darmstadt, T. (n.d.)
Basa, D., & Schneider, A. (2015)
Bimanual robot skills: MP encoding, dimensionality reduction and reinforcement learning
Colomé, A. (2017) · UPCommons institutional repository (Universitat Politècnica de Catalunya) · doi:10.5821/dissertation-2117-119092
Team Report and Code Release 2015
Laue, T., Richter-Klug, J., Stiensmeier, J., Stolpmann, A., & Thielke, F. (2015)
Action Generation Adapted to Low-Level and High-Level Robot-Object Interaction States
Maestre, C., Mukhtar, G., Gonzales, C., & Doncieux, S. (2019) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2019.00056
Colomé, A., & Carme Torras, A. (2019) · Springer tracts in advanced robotics · doi:10.1007/978-3-030-26326-3_1
Learning to control planar hitting motions in a minigolf-like task
Kronander, K., Khansari-Zadeh, M. S. M., & Billard, A. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6048048
Movement Primitive Libraries Learning for Industrial Manipulation Tasks
Xue, A., Li, X., Liu, C., & Cao, X. (2021) · Communications in computer and information science · doi:10.1007/978-981-16-2336-3_26
A highly efficient motion generation framework with error tolerance for table tennis robot
Wang, Y., Sun, Z., Zhang, H., Zhang, W., He, Q., Zhang, Q., Cai, B., & Song, B. (2025) · The International Journal of Robotics Research · doi:10.1177/02783649251396979
“For example, Kober et al. (2010) reformulated this framework to enable the racket to generate a desired hitting point and velocity.”
Rouxel, Q. (2017) · theses.fr (ABES) · doi:10.70675/d62be92fz517ez4fe8z9c8cz67be1bf2144c · Cited by 1
Quaternion DMP with Controllable Final Angular Velocity for Robot Skill Generalization
Yao, X., Zhang, E., Liao, W., & Shen, Y. (2026) · Electronics · doi:10.3390/electronics15102085
Xu, Z., Zou, Y., & Zhang, J. (2026) · Signal Image and Video Processing · doi:10.1007/s11760-026-05379-2
“These approaches typically involve two stages: trajectory prediction from human demonstrations and subsequent trajectory modulation [18].”
Metadata
- Authors
- Kober, J., Mülling, K., Krömer, O., Lampert, C. H., Schölkopf, B., & Peters, J.
- Affiliations
- Max Planck Institute for Biological Cybernetics (DE)
- Venue
- 2010 IEEE International Conference on Robotics and Automation
- Publisher
- IEEE
- Published
- 2010-05
- Pages
- 853-858
- ISSN
- 1050-4729 (ISSN-L)
- OpenAlex
- W1982803779
- Cited by
- 152
- References
- 21
- Citations by year
- 2026: 2 · 2025: 4 · 2024: 2 · 2023: 6 · 2022: 8 · 2021: 11 · 2020: 9 · 2019: 13 · 2018: 9 · 2017: 12 · 2016: 9 · 2015: 13 · 2014: 17 · 2013: 14 · 2012: 8
- FWCI
- 21.97
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Proceedings - IEEE International Conference on Robotics and Automation/Proceedings (publishedVersion)
- TUbilio (Technical University of Darmstadt) (submittedVersion)
- Max Planck Institute for Plasma Physics (submittedVersion)
- MPG.PuRe (Max Planck Society) (submittedVersion)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Motor Control and Adaptation
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Engineering › Control and Systems Engineering › Robot Manipulation and Learning
- Keywords
- motor learning, dynamical systems, motor primitives, visual servoing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.