Resilient Machines Through Continuous Self-Modeling
The DOI matches its Crossref record; 31 of the 31 references it cites were found in a registry or an index.
Abstract
Animals sustain the ability to operate after injury by creating qualitatively different compensatory behaviors. Although such robustness would be desirable in engineered systems, most machines fail in the face of unexpected damage. We describe a robot that can recover from such change autonomously, through continuous self-modeling. A four-legged machine uses actuation-sensation relationships to indirectly infer its own structure, and it then uses this self-model to generate forward locomotion. When a leg part is removed, it adapts the self-models, leading to the generation of alternative gaits. This concept may help develop more robust machines and shed light on self-modeling in animals.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Bongard et al. (2006) have created an artificial ''starfish'' that gradually develops an explicit internal self-model.”
From Empirical perspectives from the self-model theory of subjectivity: a brief summary with examples (2007) · cites this paper as Bongard et al. (2006)
In the citing paper
“When part of its leg is removed, it adapts its self-model and generates alternative gaits -it learns to limp.”
From Empirical perspectives from the self-model theory of subjectivity: a brief summary with examples (2007) · cites this paper as Bongard et al. (2006)
In the citing paper
“Their four-legged machine uses actuation-sensation relationships to indirectly infer its own structure and then uses this self-model to generate forward locomotion.”
From Empirical perspectives from the self-model theory of subjectivity: a brief summary with examples (2007) · cites this paper as Bongard et al. (2006)
Citations
Cites
System Identification: Theory for the User
Lennart Ljung, A. (1987) · Cited by 17,690
Thrun, S. (2002) · Communications of the ACM · doi:10.1145/504729.504754 · Cited by 7,994
(1999) · Choice Reviews Online · doi:10.5860/choice.36-5109 · Cited by 2,528
(1987) · Medical Entomology and Zoology · Cited by 2,512
Efficient Bipedal Robots Based on Passive-Dynamic Walkers
Collins, S., Ruina, A., Tedrake, R., & Wisse, M. (2005) · Science · doi:10.1126/science.1107799 · Cited by 1,873
Modelling and Control of Robot Manipulators
Sciavicco, L., & Siciliano, B. (2000) · Advanced Textbooks in Control and Signal Processing · doi:10.1007/978-1-4471-0449-0 · Cited by 1,701
Introduction to Machine Learning
(2019) · Series in Machine Perception and Artificial Intelligence · doi:10.1142/9789811201967_0001 · Cited by 1,639
Evolutionary Robotics: The Biology, Intelligence, and Technology of Self-Organizing Machines
Nolfi, S., & Floreano, D. (2000) · The MIT Press eBooks · doi:10.7551/mitpress/2889.001.0001 · Cited by 1,513
Crick, F., & Koch, C. (2003) · Nature Neuroscience · doi:10.1038/nn0203-119 · Cited by 1,492
RHex: A Simple and Highly Mobile Hexapod Robot
Saranli, U., Buehler, M., & Koditschek, D. E. (2001) · The International Journal of Robotics Research · doi:10.1177/02783640122067570 · Cited by 1,322
Research works printed but not linked: 3
- H. S. Seung, M. Opper, H. Sompolinsky, in Proceedings of the 5th Workshop on Computational Learning Theory (ACM Press, New York, 1992), pp. 287–294.
- J. Bongard, H. Lipson, J. Mach. Learn. Res.6, 1651 (2005).
- R. A. Brooks, in Proceedings of the 1st European Conference on Artificial Life, F. J. Varela, P. Bourgine, Eds. (Springer-Verlag, Berlin, 1992), pp. 3–10.
Cited by
Items 201–300 of 861
Evolutionary multiobjective design of a flexible caudal fin for robotic fish
Clark, A. J., Tan, X., & McKinley, P. K. (2015) · Bioinspiration & Biomimetics · doi:10.1088/1748-3190/10/6/065006 · Cited by 31
The role of attention in robot self-awareness
Novianto, R., & Williams, M. (2009) · IEEE RO-MAN · doi:10.1109/roman.2009.5326155 · Cited by 31
“body model to match to the best of its environment [3].”
BackgroundInfluentialUnderstanding the sense of self through robotics
Prescott, T. J., Vogeley, K. T., & Wykowska, A. (2024) · Science Robotics · doi:10.1126/scirobotics.adn2733 · Cited by 31
Automatic gait generation in modular robots: “to oscillate or to rotate; that is the question”
Pouya, S., van den Kieboom, J., Badri–Spröwitz, A., & Ijspeert, A. J. (2010) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2010.5649025 · Cited by 30
“These methods usually use predefined sequences but can be adjusted during runtime [12].”
MethodShreesha, L., & Levin, M. G. (2023) · Entropy · doi:10.3390/e25010131 · Cited by 30
Learning agent’s spatial configuration from sensorimotor invariants
Laflaquière, A., O'Regan, J. K., Argentieri, S., Gas, B., & Terekhov, A. V. (2015) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2015.01.003 · Cited by 31
“ed 1:75 units in front of the agent. The analysis is restricted to 2500 dierent arm congurations. Figure Fig 5a presents the working space and a sampling of the 2500 arm congurations. 11 j i i[2] j[7] min l(g( , )i[2] j[l] ) min k(g( , )i[k] j[7] ) ( , )i j Figure 4: The Hausdor distance between two sets Mi and Mj cor…”
BackgroundInfluentialCircadian hormone control in a human-on-a-chip: In vitro biology’s ignored component?
Cyr, K. J., Avaldi, O. M., & Wikswo, J. P. (2017) · Experimental Biology and Medicine · doi:10.1177/1535370217732766 · Cited by 30
Toward an Integrated Approach to Perception and Action: Conference Report and Future Directions
Gordon, G., Kaplan, D. M., Lankow, B. S., Little, D. Y.-J., Sherwin, J. S., Suter, B. A., & Thaler, L. (2011) · Frontiers in Systems Neuroscience · doi:10.3389/fnsys.2011.00020 · Cited by 30
“tor learning, computation, cognitive development, and biological evolution (Bongard et al., 2006).”
BackgroundThe Importance of Being Active
Brembs, B. (2009) · Journal of Neurogenetics · doi:10.1080/01677060802471643 · Cited by 30
“Even robots can use such ‘‘self-modeling’’ to generate a continuously updated model of themselves and their environment (Bongard et al., 2006).”
BackgroundRubenstein, M., Sai, Y., Chuong, C., & Shen, W.-M. (2009) · The International Journal of Developmental Biology · doi:10.1387/ijdb.092937mr · Cited by 30
Deep kinematic inference affords efficient and scalable control of bodily movements
Priorelli, M., Pezzulo, G., & Stoianov, I. P. (2023) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2309058120 · Cited by 29
“Third, it affords self-modeling [38] by continuously inferring the kinematic”
BackgroundFault-tolerant gait learning and morphology optimization of a polymorphic walking robot
Christensen, D. J., Larsen, J. C., & Støy, K. (2013) · Evolving Systems · doi:10.1007/s12530-013-9088-3 · Cited by 29
“In addition, the robot could learn a self-model (Bongard et al 2006) which in some situations could be used to predict and avoid falling over.”
BackgroundMethodInfluentialSchilling, M., Paskarbeit, J., Schmitz, J., Schneider, A., & Cruse, H. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6385709 · Cited by 29
“Other approaches try to come up with a body schema starting from scratch (an impressive example is the work by Bongard [2] in which a starfish like robot synthesizes a body scheme and a simple walking behavior and can adapt both even when one leg of the robot is shortened).”
BackgroundInfluentialChaotic Exploration and Learning of Locomotion Behaviors
Shim, Y., & Husbands, P. (2012) · Neural Computation · doi:10.1162/neco_a_00313 · Cited by 29
Learning Fast Quadruped Robot Gaits with the RL PoWER Spline Parameterization
Shen, H., Yosinski, J., Kormushev, P. S., Caldwell, D. G., & Lipson, H. (2012) · Cybernetics and Information Technologies · doi:10.2478/cait-2012-0022 · Cited by 29
Self-citationAdaptive Body Scheme Models for Robust Robotic Manipulation
Sturm, J., Plagemann, C., & Burgard, W. (2008) · Robotics: Science and Systems · doi:10.15607/rss.2008.iv.015 · Cited by 29
Enders, J. R., Marasco, C. C., Kole, A., Nguyen, B. H., Sevugarajan, S., Seale, K. T., Wikswo, J. P., & MCLEAN, J. A. J. (2010) · IET Systems Biology · doi:10.1049/iet-syb.2010.0012 · Cited by 28
Evolvability signatures of generative encodings: Beyond standard performance benchmarks
Tarapore, D., & Mouret, J.-B. (2015) · Information Sciences · doi:10.1016/j.ins.2015.03.046 · Cited by 29
Predictable internal brain dynamics in EEG and its relation to conscious states
Yoo, J., Kwon, J., & Choe, Y. (2014) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2014.00018 · Cited by 28
“Due to the non-linear property, compared to linear predictors such as autoregressive models (Blinowska and Malinowski, 1991), neural networks are known to give generally better performance (Coyle et al., 2005).”
BackgroundTowards the targeted environment-specific evolution of robot components
Collins, J., Geles, W., Howard, D., & Maire, F. (2018) · Genetic and Evolutionary Computation Conference (GECCO) · doi:10.1145/3205455.3205541 · Cited by 28
“internal models [6, 9], however it is unknown if model fidelity can match complex real-world environmental conditions.”
BackgroundEmbodied AI beyond Embodied Cognition and Enactivism
Manzotti, R. (2019) · Philosophies · doi:10.3390/philosophies4030039 · Cited by 28
QED: Using Quality-Environment-Diversity to Evolve Resilient Robot Swarms
Bossens, D. M., & Tarapore, D. (2020) · IEEE Transactions on Evolutionary Computation · doi:10.1109/tevc.2020.3036578 · Cited by 28
Gait optimization for roombots modular robots — Matching simulation and reality
Möckel, R., Perov, Y., Nguyen, A., Vespignani, M., Bonardi, S., Pouya, S., Badri–Spröwitz, A., van den Kieboom, J., Wilhelm, F., & Ijspeert, A. J. (2013) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2013.6696820 · Cited by 27
“The parameter space exploration requires experiments running in real time and unless many robots with wellmatched behavior are available the optimization process cannot be parallelized.”
BackgroundKim, K.-J., & Cho, S. (2011) · Applied Soft Computing · doi:10.1016/j.asoc.2011.12.002 · Cited by 27
“Finally, unlike the two passive approaches, researchers try to build active fault-tolerant systems which reconfigure themselves when faults occur [5].”
Online tree-based planning for active spacecraft fault estimation and collision avoidance
Ragan, J., Rivière, B., Hadaegh, F. Y., & Chung, S. (2024) · Science Robotics · doi:10.1126/scirobotics.adn4722 · Cited by 26
“These approaches to fault estimation are similar to the field of robotic self-modeling ( 32 ), which has recently used exploratory actions to distinguish between possible dynamic models of robotic hands ( 33 ) and to learn visual self-models ( 34 ).”
MethodThe Convergence of AI and animal-inspired robots for ecological conservation
Afzal, N., Rehman, M. U., Seneviratne, L. D., & Hussain, I. (2024) · Ecological Informatics · doi:10.1016/j.ecoinf.2024.102950 · Cited by 26
Evolution of a predictive internal model in an embodied and situated agent
Gigliotta, O., Pezzulo, G., & Nolfi, S. (2011) · Theory in Biosciences · doi:10.1007/s12064-011-0128-x · Cited by 26
Woodford, G. W., Pretorius, C. J., & du Plessis, M. C. (2015) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2015.10.011 · Cited by 26
“Evolving a robot’s morphology usually necessitates changing the model of what the robot looks like and evolving the physics model parameters of the simulator [14, 18].”
BackgroundCurious Replay for Model-based AdaptationPreprint
Kauvar, I., Doyle, C., Zhou, L., & Haber, N. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.15934 · Cited by 28
“Approaches toward adaptation include using system identification (Bongard et al., 2006; Cully et al., 2015; Kumar et al., 2021), metalearning (Finn et al.”
BackgroundMethodSahimi, M. (2024) · Physical review. E · doi:10.1103/physreve.109.041001 · Cited by 25
Encouraging reactivity to create robust machines
Lehman, J., Risi, S., D’Ambrosio, D., & Stanley, K. O. (2013) · Adaptive Behavior · doi:10.1177/1059712313487390 · Cited by 25
“In this way, it is possible to co-evolve simulators and controllers to reduce discrepancies (Bongard & Lipson, 2004; Zagal & Ruiz-DelSolar, 2007; Bongard et al., 2006), or to penalize behaviors that exploit them (Koos et al., 2012).”
BackgroundInfluentialNaudé, W. A. (2019) · SSRN Electronic Journal · doi:10.2139/ssrn.3390207 · Cited by 25
“See for instance the Brynjolfsson and McAfee (2012) book Race Against the Machine or the Rise of the Robots by Ford (2016).”
BackgroundPhotomorphogenesis for robot self-assembly: adaptivity, collective decision-making, and self-repair
Soorati, M. D., Heinrich, M. K., Ghofrani, J., Zahadat, P., & Hamann, H. (2019) · Bioinspiration & Biomimetics · doi:10.1088/1748-3190/ab2958 · Cited by 24
“Biological morphogenesis in both plants and animals has inspired advances across robotics [56 – 59], including approaches to self-assembly [27, 60].”
BackgroundTutorial Review of Bio-Inspired Approaches to Robotic Manipulation for Space Debris Salvage
Ellery, A. A. (2020) · Biomimetics · doi:10.3390/biomimetics5020019 · Cited by 24
“Continuous self-modelling by inferring its own structure from sensorimotor causal relationships provides the ability to engage in autonomous compensatory behaviours due to injury [45].”
BackgroundInfluentialIn Silicon No One Can Hear You Scream: Evolving Fighting Creatures
Miconi, T. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-78671-9_3 · Cited by 24
“Simultaneously, Bongard and colleagues [5] have explored new directions in the joint evolution of morphology and behaviour: actual robots in the real world engage in continuous self-modelling and self-simulation, in effect evolving models of themselves.”
BackgroundAracna: An Open-Source Quadruped Platform for Evolutionary Robotics
Lohmann, S., Yosinski, J., Gold, E., Clune, J., Blum, J., & Lipson, H. (2012) · Conference on Artificial Life (ALIFE) · doi:10.1162/978-0-262-31050-5-ch051 · Cited by 24
Self-citationResilient machines through adaptive morphology
Häuser, H. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0076-6 · Cited by 24
“Previous work suggested purely algorithmic approaches to deal with damage by providing an alternative control strategy for the impaired body (2,3).”
Evolved Machines Shed Light on Robustness and Resilience
Bongard, J., & Lipson, H. (2014) · Proceedings of the IEEE · doi:10.1109/jproc.2014.2312844 · Cited by 23
“In [7], no such separation was made: forward models were constrained to actuatable kinematic chains.”
MethodSelf-citationAutomatic system identification based on coevolution of models and tests
Koos, S., Mouret, J.-B., & Doncieux, S. (2009) · IEEE Congress on Evolutionary Computation (CEC) · doi:10.1109/cec.2009.4982995 · Cited by 23
“The paper ends with some conclusions in section VIII.”
BackgroundAI AwarenessPreprint
Li, X., Shi, H., Xu, R., & Xu, W. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2504.20084 · Cited by 23
Meta Reinforcement Learning of Locomotion Policy for Quadruped Robots With Motor Stuck
Chen, C., Li, C., Lu, H., Wang, Y., & Xiong, R. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2024.3424328 · Cited by 22
“Hence, it is crucial to develop an algorithm that can equip quadruped robots with fault tolerance [3], [4], [5], [6].”
Jones, S., Studley, M., Hauert, S., & Winfield, A. F. T. (2018) · Frontiers in Robotics and AI · doi:10.3389/frobt.2018.00011 · Cited by 22
“Usinginternalsimulation,modelscanbemeansofdetecting malfunctionandadapting(Bongardetal.,2006),orasking what-if questions, so as to evaluate the consequences of possible actions in simulation (Marques and Holland, 2009).”
BackgroundPersonogenesis Through Imitating Human Behavior in a Humanoid Robot “Alter3”
Masumori, A., Maruyama, N., & Ikegami, T. (2021) · Frontiers in Robotics and AI · doi:10.3389/frobt.2020.532375 · Cited by 22
“A more pertinent study is that of the self-modeling agent proposed by Bongard et al. (2006).”
BackgroundArtificial consciousness: A discipline between technological and theoretical obstacles
Manzotti, R., & Tagliasco, V. (2008) · Artificial Intelligence in Medicine · doi:10.1016/j.artmed.2008.07.002 · Cited by 22
“Another example is offered by the four-legged Starfish robot, built by Josh Bongard, Victor Zykov and Hod Lipson, that uses the readings from its sensors to create a computer model of its own structure and movement [44].”
BackgroundMethodInfluentialCrossing the Reality Gap: a Short Introduction to the Transferability ApproachPreprint
Mouret, J.-B., Koos, S., & Doncieux, S. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1307.1870 · Cited by 22
“An attractive idea is to automatically design a simulator, for instance by learning a surrogate model of the fitness function (Jin, 2005), or, following a related idea, to automatically improve an existing simulator (Bongard et al., 2006; Zagal and Ruiz-Del-Solar, 2007).”
BackgroundInfluentialTeaching robots to build simulations of themselves
Hu, Y., Lin, J., & Lipson, H. (2025) · Nature Machine Intelligence · doi:10.1038/s42256-025-01006-w · Cited by 21
“Robots should learn to create models for themselves that not only can generate motor commands to solve the problems but also can adapt to changes ang recovery from damage 32 .”
Self-citationConservation of Information: Reverse engineering dark social systems
Lawless, W. F., Rifkin, S., Sofge, D., Hobbs, S. H., Angjellari-Dajci, F., Chaudron, L., & Wood, J. (2010) · Structure and Dynamics eJournal of Anthropological and Related Sciences · doi:10.5070/sd942003309 · Cited by 21
Interoceptive robustness through environment-mediated morphological development
Kriegman, S., Cheney, N., Corucci, F., & Bongard, J. (2018) · Genetic and Evolutionary Computation Conference (GECCO) · doi:10.1145/3205455.3205529 · Cited by 21
“Although generally absent from engineered systems (but see [3, 11]), the canonical form of robustness is seen to some extent in all organisms, and it comes from the act of development itself.”
BackgroundSelf-citationCounterfactual Explanation and Causal Inference In Service of Robustness in Robot Control
Smith, S. C., & Ramamoorthy, S. (2020) · IEEE International Conference on Development and Learning (ICDL) · doi:10.1109/icdl-epirob48136.2020.9278061 · Cited by 21
“The generation of counterfactuals is analogous to adaptive methods in configuration space [45].”
MethodSvensson, H., Thill, S., & Ziemke, T. (2013) · Adaptive Behavior · doi:10.1177/1059712313491295 · Cited by 21
Egocentric visual self-modeling for autonomous robot dynamics prediction and adaptation
Hu, Y., Chen, B., & Lipson, H. (2025) · npj Robotics · doi:10.1038/s44182-025-00031-6 · Cited by 20
“Self-modeling is a form of model-based control that decouples (deconvolves) a model of the robot (the “self-model”) from a model of its environment and task [10], [11].”
Self-citationPugach, G., Pitti, A., & Gaussier, P. (2015) · Advanced Robotics · doi:10.1080/01691864.2015.1092395 · Cited by 20
“[10] In the context of haptic devices, various types of sensors have been proposed to transform tactile information into electrical signals,[11–13] but not all of them are good for dynamic adaptation when the structure of the sensory information changes.”
Multi-objective evolution of fast and stable gaits on a physical quadruped robotic platform
Nygaard, T. F., Tørresen, J., & Glette, K. (2016) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci.2016.7850167 · Cited by 20
“This has resulted in many mechanically simple and fairly limited robots being used in current evolutionary experiments [4], [7], [10], though we are starting to see more capable robots emerging from the traditional robotics field [11], [12] which might serve as evolutionary platforms for real world…”
BackgroundAmbient intelligence in self-organising assembly systems using the chemical reaction model
Frei, R., Serugendo, G. D. M., & Şerbănuţă, T. F. (2010) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-010-0016-0 · Cited by 20
“In the context of computer science and robotics (Bongard et al. 2006; Di Marzo Serugendo et al. 2007) and self-organising systems (Di Marzo Serugendo 2009), resilience means ‘dependability when facing changes’ (Laprie 2008), or in other words, its ability to maintain dependability while…”
BackgroundA distributed strategy for gait adaptation in modular robots
Christensen, D. J., Schultz, U. P., & Støy, K. (2010) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/robot.2010.5509942 · Cited by 19
“demonstrated learning of locomotion and adaptation to changes in the configuration of a modular robot [1].”
BackgroundWhat’s Next: Recruitment of a Grounded Predictive Body Model for Planning a Robot’s Actions
Schilling, M., & Cruse, H. (2012) · Frontiers in Psychology · doi:10.3389/fpsyg.2012.00383 · Cited by 19
“Another approach related to ours is the work of Bongard et al. (2006). These authors have used an internal model of the body in a starfish-like four-legged robot. In their system, the internal model was used in internal simulation loops to evolve locomotion controllers. The internal model was used to predict sensory c…”
BackgroundInfluentialTIME, CONSCIOUSNESS, AND MIND UPLOADING
Choe, Y., Kwon, J., & Chung, J. R. (2012) · International Journal of Machine Consciousness · doi:10.1142/s179384301240015x · Cited by 19
“On the other hand, Bongard et al. [2006] showed that through the use of an internal self model, physical sensorimotor agents can show resilient behavior when part of the agent becomes damaged (such as amputated limbs, etc.). There are more instances of prediction being detected in brain function. Rao & Ballard [1999]…”
BackgroundInfluentialNeuroevolution of Inverted Pendulum Control: A Comparative Study of Simulation Techniques
Pretorius, C. J., du Plessis, M. C., & Gonsalves, J. W. (2017) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-017-0465-1 · Cited by 19
“Bongard, Zykov and Lipson [8], for example, showed that robust robotic simulators could be constructed by comparing behaviours of a real-world robot to those predicted by a simulator, even in the case where the robot’s morphology underwent dramatic changes.”
BackgroundGonzález, A. G., García-Córdova, F., Ortiz, F. J., Alonso, D., & Gilabert, J. (2016) · Autonomous Robots · doi:10.1007/s10514-016-9602-0 · Cited by 19
“This capacity gives them resilience, and a resilient robot can develop new settings through continuous self-modeling (Bongard et al. 2006; Sun et al. 2011).”
BackgroundObserving the Effects of Overdesign in the Automatic Design of Control Software for Robot Swarms
Birattari, M., Delhaisse, B., Francesca, G., & Kerdoncuff, Y. (2016) · Lecture notes in computer science · doi:10.1007/978-3-319-44427-7_13 · Cited by 19
“” A number of ideas have been proposed to reduce the impact of the reality gap, including methods to increase the realism of simulation [31,36] and design protocols that alternate simulation with runs in reality [5,33].”
BackgroundReaCog, a Minimal Cognitive Controller Based on Recruitment of Reactive Systems
Schilling, M., & Cruse, H. (2017) · Frontiers in Neurorobotics · doi:10.3389/fnbot.2017.00003 · Cited by 20
“With respect to reaCog the most similar approach has been pursued by Bongard et al. (2006).”
BackgroundMethodInfluentialAnalysis of Lamarckian evolution in morphologically evolving robots
Jelisavcic, M., Kiesel, R., Glette, K., Haasdijk, E., & Eiben, A. E. (2017) · European Conference on Artificial Life · doi:10.1162/isal_a_038 · Cited by 19
Evolving the Behavior of Machines: From Micro to Macroevolution
Mouret, J.-B. (2020) · iScience · doi:10.1016/j.isci.2020.101731 · Cited by 19
“…al., 2015; Cully et al., 2015); improve the simulator either with extensive tuning (Tan et al., 2018) or by using data to improve it automatically (Bongard et al., 2006), which is called ‘‘system identification’’ in robotics; Make the policies (controllers) more robust to different dynamics by: –…”
BackgroundSelf discovery enables robot social cognition: Are you my teacher?
Kaipa, K. N., Bongard, J., & Meltzoff, A. N. (2010) · Neural Networks · doi:10.1016/j.neunet.2010.07.009 · Cited by 19
Self-citationAIT Foundations of Structured Experience
Ruffini, G., & Lopez-Sola, E. (2022) · Journal of Artificial Intelligence and Consciousness · doi:10.1142/s2705078522500047 · Cited by 20
Chattunyakit, S., Kobayashi, Y., Emaru, T., & Ravankar, A. A. (2019) · Applied Sciences · doi:10.3390/app9040799 · Cited by 18
Sim2real gap is non-monotonic with robot complexity for morphology-in-the-loop flapping wing design
Rosser, K., Kok, J. M., Chahl, J., & Bongard, J. (2020) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra40945.2020.9196539 · Cited by 18
“Bongard et al [3] successfully crossed the reality gap by developing a method for continuous morphology self-modelling while operating on the physical robot and co-evolving a set of controllers to suit in simulation.”
MethodSelf-citationCoevolutionary Modelling of a Miniature Rotorcraft.
Holland, O. E., & Nardi, R. D. (2008) · Figshare · Cited by 18
A coevolutionary approach to learn animal behavior through controlled interaction
Li, W., Gauci, M., & Groß, R. (2013) · Genetic and Evolutionary Computation Conference (GECCO) · doi:10.1145/2463372.2465801 · Cited by 18
“[5] study the problem how a robot can infer its own morphology through a process of continuous selfmodeling.”
BackgroundTransferring spatial perception between robots operating in a shared workspace
Leitner, J., Harding, S., Frank, M., Förster, A., & Schmidhuber, J. (2012) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2012.6385642 · Cited by 18
“…Molle Institute for Artificial Intelligence (IDSIA), the Università della Svizzera italiana (USI) & the Scuola universitaria professionale della Svizzera italiana (SUPSI) juxi@idsia.ch used to position an object in the shared workspace to provide the humanoid with the information to learn from.”
BackgroundA survey of evolutionary and embryogenic approaches to autonomic networking
Miorandi, D., Yamamoto, L., & Pellegrini, F. D. (2009) · Computer Networks · doi:10.1016/j.comnet.2009.08.021 · Cited by 18
Optimal Experiment Design for Coevolutionary Active Learning
Ly, D., & Lipson, H. (2013) · IEEE Transactions on Evolutionary Computation · doi:10.1109/tevc.2013.2281529 · Cited by 18
“Flavors of this coevolution active learning have been used in fields such as robotics [1] and automated science [2], [3].”
MethodSelf-citationNeuromodulation Improves the Evolution of Forward Models
Norouzzadeh, M. S., & Clune, J. (2016) · Genetic and Evolutionary Computation Conference (GECCO) · doi:10.1145/2908812.2908837 · Cited by 18
“Being able to predict the consequences of actions could grant robots the ability to better adapt to changes to their environments or bodies; specifically, they could mentally simulate possible movements to select an appropriate one without having to incur the costs and hazards of trying each action the real world [3,1…”
BackgroundMemetic robot control evolution and adaption to reality
Ruud, E.-L. M., Samuelsen, E., & Glette, K. (2016) · IEEE Symposium Series on Computational Intelligence (SSCI) · doi:10.1109/ssci.2016.7850169 · Cited by 18
“through real world sampling [19], [20], [21], or concentrating around behaviors which are known to match well in both simulation and reality [22].”
BackgroundToward Robot Self-Consciousness (II): Brain-Inspired Robot Bodily Self Model for Self-Recognition
Zeng, Y., Zhao, Y., Bai, J., & Xu, B. (2017) · Cognitive Computation · doi:10.1007/s12559-017-9505-1 · Cited by 18
“provide an architecture for continuous self-modeling [16].”
Mastering Uncertainty in Mechanical Engineering
Pelz, P. F., Groche, P., Pfetsch, M. E., & Schäffner, M. F. (2023) · Technischen Universität Darmstadt · doi:10.26083/tuprints-00023754 · Cited by 18
Autonomous Maintenance for Through-Life Engineering
Farnsworth, M., Bell, C., Khan, S., & Tomiyama, T. (2014) · Decision engineering · doi:10.1007/978-3-319-12111-6_23 · Cited by 19
Policy Stitching: Learning Transferable Robot PoliciesPreprint
Jian, P., Lee, E., Bell, Z. I., Zavlanos, M. M., & Chen, B. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2309.13753 · Cited by 18
“Model-based robot learning, in particular self-modeling or self-identification learning [2, 3, 4, 5, 6, 7, 8, 9, 10], aims to learn a predictive model of the robot’s kinematics and dynamics and then employ this model for various downstream tasks through model predictive control.”
MethodSelf Exploration of Sensorimotor Spaces in Robots.
Benureau, F. (2015) · theses.fr (ABES) · doi:10.70675/1d482feaz6e01z4426zbc1cz25388edcb189 · Cited by 18
What Matters in Constructing a Visual Servoing Scheme: A Review of Key Issues and Solutions
Xu, F., Wang, H., & Gao, H. (2025) · IEEE/ASME Transactions on Mechatronics · doi:10.1109/tmech.2025.3605943 · Cited by 17
“Such a problem has long been highlighted as challenging since models can vary due to diverse factors, requiring numerous equations to accurately describe the system [141].”
Learning realistic lip motions for humanoid face robots
Hu, Y., Lin, J., Goldfeder, J., Wyder, P. M., Cao, Y., Tian, S., Wang, Y., Wang, J., Wang, M., Zeng, J., et al. (2026) · Science Robotics · doi:10.1126/scirobotics.adx3017 · Cited by 17
“To bridge this complexity, we used a learned self-model ( 27 – 30 ).”
Self-citationAI in the 21st Century – With Historical Reflections
Lungarella, M., Iida, F., Bongard, J., & Pfeifer, R. (2007) · Springer eBooks · doi:10.1007/978-3-540-77296-5_1 · Cited by 18
“Mechanisms need to be created for the embodied and embedded systems to collect relevant learning material on their own [ 36 ] and for learning to take place in an “ecological context” (i.e.”
BackgroundSelf-citationFrom Adaptive Locomotion to Predictive Action Selection – Cognitive Control for a Six-Legged Walker
Schilling, M., Paskarbeit, J., Ritter, H. J., Schneider, A., & Cruse, H. (2021) · IEEE Transactions on Robotics · doi:10.1109/tro.2021.3106832 · Cited by 17
Physical Neural Cellular Automata for 2D Shape Classification
Walker, K. L., Palm, R. B., Garcia, R. M., Faíña, A., Støy, K., & Risi, S. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981214 · Cited by 17
“ability to self-classify with the ability to consider outcomes of multiple future actions through self-modeling [5], [4] could further increase the potential application areas of these robotic systems.”
BackgroundSafeAPT: Safe Simulation-to-Real Robot Learning Using Diverse Policies Learned in Simulation
Kaushik, R., Arndt, K., & Kyrki, V. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3177294 · Cited by 17
Evolutionary Humanoid Robotics: Past, Present and Future
Eaton, M. (2007) · Springer eBooks · doi:10.1007/978-3-540-77296-5_5 · Cited by 17
A New Approach to Organizations: Stability and Transformation in Dark Social Networks
Lawless, W. F., Angjellari-Dajci, F., Sofge, D., Grayson, J. M., Sousa, J., & Rychly, L. (2011) · Journal of Enterprise Transformation · doi:10.1080/19488289.2011.623029 · Cited by 17
“…influences composed of two or more simultaneous (bistable) values that linearly combine under constructive interference into a single worldview (consensus), or under destructive interference to disambiguate dissimilar views into the best concrete plan to execute (Bongard, Zykov, and Lipson, 2006).”
BackgroundTannenbaum, E. (2008) · Consciousness and Cognition · doi:10.1016/j.concog.2008.06.003 · Cited by 17
The distributed co-evolution of an embodied simulator and controller for swarm robot behaviours
O’Dowd, P., Winfield, A. F. T., & Studley, M. (2011) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2011.6094600 · Cited by 17
“[2] develop the ‘estimation/exploration’ algorithm which”
Innocent Until Proven Guilty: Reducing Robot Shaping From Polynomial to Linear Time
Bongard, J. (2011) · IEEE Transactions on Evolutionary Computation · doi:10.1109/tevc.2010.2096540 · Cited by 17
Self-citationConcepts of Self-Repairing Systems
Bell, C., McWilliam, R. P., Purvis, A., & Tiwari, A. (2013) · Measurement and Control · doi:10.1177/0020294013492285 · Cited by 17
Towards a simple robotic theory of mind
Kim, K.-J., & Lipson, H. (2009) · Workshop on Performance Metrics for Intelligent Systems (PerMIS) · doi:10.1145/1865909.1865937 · Cited by 17
“The inference is based on the exploration-estimation algorithm (EEA) used in reverse engineering of nonlinear-dynamical systems [9] and robot’s self modeling [10].”
MethodSelf-citationSilva, J. F., Correia, L., & Christensen, A. (2016) · Scholarpedia · doi:10.4249/scholarpedia.33333 · Cited by 17
The road to everywhere : Evolution, complexity and progress in natural and artificial systems
Miconi, T. (2008) · University of Birmingham Institutional Research Archive (University of Birmingham) · Cited by 17
The distributed co-evolution of an on-board simulator and controller for swarm robot behaviours
O’Dowd, P., Studley, M., & Winfield, A. F. T. (2014) · Evolutionary Intelligence · doi:10.1007/s12065-014-0112-8 · Cited by 16
“The work of Bongard et al. [2] is a primary example of a robot that is able to adapt a self-model of morphology.”
BackgroundMethodInfluentialWoodford, G. W., du Plessis, M. C., & Pretorius, C. J. (2016) · Robotics and Autonomous Systems · doi:10.1016/j.robot.2016.11.018 · Cited by 16
“Optimisation approaches such as Any-time Learning [9], Estimation-Exploration Algorithm [10] and Back to Reality [6] are actively involved in the improvement of the simulator model’s parameters.”
BackgroundMethodAchieving natural behavior in a robot using neurally inspired hierarchical perceptual control
Barter, J. W., & Yin, H. H. (2021) · iScience · doi:10.1016/j.isci.2021.102948 · Cited by 16
“Unlike other robot control architectures that performmodel-based control and planning (Bledt et al., 2018; Bongard et al., 2006; Focchi et al., 2017; Hwangbo et al., 2019; Tsounis et al., 2020), our design generates robust and adaptive goal-directed behavior without requiring a model of the…”
BackgroundHow to Train Your Drone: Exploring the umwelt as a design metaphor for human-drone interaction
Delfa, J. L., Garrett, R., Lampinen, A., & Höök, K. (2024) · Designing Interactive Systems Conference · doi:10.1145/3643834.3660737 · Cited by 16
Learning Spatial Object Localization from Vision on a Humanoid Robot
Leitner, J., Harding, S., Frank, M., Förster, A., & Schmidhuber, J. (2012) · International Journal of Advanced Robotic Systems · doi:10.5772/54657 · Cited by 16
“For other robotic platforms, machine learning has been used to estimate the kinematic model, for example, Bongard et al. employed sensory feedback to learn [12], but without high-dimensional sensory information such as camera images.”
MethodAutoFac: The Perpetual Robot Machine
Nitschke, G., & Howard, D. (2021) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2021.3104789 · Cited by 16
Metadata
- Authors
- Bongard, J., Zykov, V., & Lipson, H.
- Affiliations
- Cornell University (US)
- Corresponding author
- Bongard, Josh
- Venue
- Science
- Published
- 2006-11-17
- Volume
- 314
- Issue
- 5802
- Pages
- 1118-1121
- ISSN
- 0036-8075 (ISSN-L)
- 1095-9203
- OpenAlex
- W2001685400
- Cited by
- 709
- References
- 31
- Citations by year
- 2026: 24 · 2025: 22 · 2024: 31 · 2023: 22 · 2022: 32 · 2021: 36 · 2020: 39 · 2019: 39 · 2018: 37 · 2017: 53 · 2016: 43 · 2015: 40 · 2014: 33 · 2013: 48 · 2012: 33
- FWCI
- 26.07
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Science (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Engineering › Control and Systems Engineering › Robotic Locomotion and Control
- Life Sciences › Neuroscience › Cognitive Neuroscience › Embodied and Extended Cognition
- Physical Sciences › Engineering › Biomedical Engineering › Prosthetics and Rehabilitation Robotics
- Keywords
- self-modeling, four-legged robot, legged locomotion, gait adaptation, structural inference, autonomous recovery, robustness
- MeSH terms
- Algorithms, Animals, Computer Simulation, Robotics
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.