Abstract
Modern manufacturing and assembly environments are characterized by a high variability in the built process which challenges human–robot cooperation. To reduce the cognitive workload of the operator, the robot should not only be able to learn from experience but also to plan and decide autonomously. Here, we present an approach based on Dynamic Neural Fields that apply brain-like computations to endow a robot with these cognitive functions. A neural integrator is used to model the gradual accumulation of sensory and other evidence as time-varying persistent activity of neural populations. The decision to act is modeled by a competitive dynamics between neural populations linked to different motor behaviors. They receive the persistent activation pattern of the integrators as input. In the first experiment, a robot learns rapidly by observation the sequential order of object transfers between an assistant and an operator to subsequently substitute the assistant in the joint task. The results show that the robot is able to proactively plan the series of handovers in the correct order. In the second experiment, a mobile robot searches at two different workbenches for a specific object to deliver it to an operator. The object may appear at the two locations in a certain time period with independent probabilities unknown to the robot. The trial-by-trial decision under uncertainty is biased by the accumulated evidence of past successes and choices. The choice behavior over a longer period reveals that the robot achieves a high search efficiency in stationary as well as dynamic environments.
Similar content being viewed by others
References
Agostini A, Torras C, Woergoetter F (2017) Efficient interactive decision-making framework for robotic applications. Artif Intell 247:187–212. https://doi.org/10.1016/j.artint.2015.04.004
Amari S (1977) Dynamics of pattern formation in lateral-inhibition type neural fields. Biol Cybern 27(2):77–87. https://doi.org/10.1007/BF00337259
Bannat A, Bautze T, Beetz M, Blume J, Diepold K, Ertelt C, Geiger F, Gmeiner T, Gyger T, Knoll A et al (2010) Artificial cognition in production systems. IEEE Trans Autom Sci Eng 8(1):148–174. https://doi.org/10.1109/TASE.2010.2053534
Bicho E, Mallet P, Schöner G (2000) Target representation on an autonomous vehicle with low-level sensors. Int J Robot Res 19(5):424–447. https://doi.org/10.1177/02783640022066950
Bicho E, Louro L, Erlhagen W (2010) Integrating verbal and nonverbal communication in a dynamic neural field architecture for human-robot interaction. Front Neurorobot 4:5. https://doi.org/10.3389/fnbot.2010.00005
Bicho E, Erlhagen W, Louro L, Costa e Silva E, Silva R, Hipolito N (2011a) A dynamic field approach to goal inference, error detection and anticipatory action selection in human-robot collaboration. New Front Human-Robot Interact. https://doi.org/10.1075/ais.2.10bic
Bicho E, Erlhagen W, Louro L, e Silva EC (2011b) Neuro-cognitive mechanisms of decision making in joint action: a human-robot interaction study. Human Mov Sci 30(5):846–868. https://doi.org/10.1016/j.humov.2010.08.012
Bicho E, Erlhagen W, Sousa E, Louro L, Hipólito N, Silva E, Silva R, Ferreira F, Machado T, Hulstijn M, et al. (2012) The power of prediction: robots that read intentions. In: 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, IEEE, pp 5458–5459, https://doi.org/10.1109/IROS.2012.6386297
Billard A, Calinon S, Dillmann R, Schaal S (2008) Robot programming by demonstration. Springer Handb Robot. https://doi.org/10.1007/978-3-540-30301-5_60
Brody CD, Hanks TD (2016) Neural underpinnings of the evidence accumulator. Curr Opin Neurobiol 37:149–157. https://doi.org/10.1016/j.conb.2016.01.003
Brody CD, Romo R, Kepecs A (2003) Basic mechanisms for graded persistent activity: discrete attractors, continuous attractors, and dynamic representations. Curr Opin Neurobiol 13(2):204–211. https://doi.org/10.1016/S0959-4388(03)00050-3
Cain N, Shea-Brown E (2012) Computational models of decision making: integration, stability, and noise. Curr Opin Neurobiol 22(6):1047–1053. https://doi.org/10.1016/j.conb.2012.04.013
Choe P, Tew JD, Tong S (2015) Effect of cognitive automation in a material handling system on manufacturing flexibility. Int J Product Econ 170:891–899. https://doi.org/10.1016/j.ijpe.2015.01.018
Coombes S (2005) Waves, bumps, and patterns in neural field theories. Biol Cybern 93(2):91–108. https://doi.org/10.1007/s00422-005-0574-y
Coombes S, beim Graben P, Potthast R, Wright J (2014) Neural fields: theory and applications. Springer, Berlin. https://doi.org/10.1007/978-3-642-54593-1
Cox BR, Krichmar JL (2009) Neuromodulation as a robot controller. IEEE Robot Autom Mag 16(3):72–80. https://doi.org/10.1109/MRA.2009.933628
Curtis CE, Lee D (2010) Beyond working memory: the role of persistent activity in decision making. Trends Cognit Sci 14(5):216–222. https://doi.org/10.1016/j.tics.2010.03.006
Erlhagen W, Bicho E (2006) The dynamic neural field approach to cognitive robotics. J Neural Eng 3:36–54. https://doi.org/10.1088/1741-2560/3/3/R02
Erlhagen W, Bicho E (2014) A dynamic field approach to natural and efficient human-robot collaboration. In: Pothast R, Weight J, Coombes S, Beim Graben P (eds) Neural fields: theory and applications. Springer, Berlin, pp 341–365. https://doi.org/10.1007/978-3-642-54593-1_13
Erlhagen W, Schöner G (2002) Dynamic field theory of movement preparation. Psychol Rev 109(3):545. https://doi.org/10.1037/0033-295X.109.3.545
Erlhagen W, Mukovskiy A, Bicho E, Panin G, Kiss C, Knoll A, Van Schie H, Bekkering H (2006) Goal-directed imitation for robots: a bio-inspired approach to action understanding and skill learning. Robot Auton Syst 54(5):353–360. https://doi.org/10.1016/j.robot.2006.01.004
Faubel C, Schöner G (2008) Learning to recognize objects on the fly: a neurally based dynamic field approach. Neural Netw 21(4):562–576. https://doi.org/10.1016/j.neunet.2008.03.007
Ferreira F, Erlhagen W, Sousa E, Louro L, Bicho E (2014) Learning a musical sequence by observation: A robotics implementation of a dynamic neural field model. In: Development and Learning and Epigenetic Robotics (ICDL-Epirob), 2014 Joint IEEE International Conferences on, pp 157–162, https://doi.org/10.1109/DEVLRN.2014.6982973
Ferreira F, Erlhagen W, Bicho E (2016) Multi-bump solutions in a neural field model with external inputs. Phys D Nonlinear Phenomen 326:32–51. https://doi.org/10.1016/j.physd.2016.01.009
Ferreira F, Wojtak W, Sousa E, Louro L, Bicho E, Erlhagen W (2020) Rapid learning of complex sequences with time constraints: a dynamic neural field model. IEEE Trans Cognit Develop Syst. https://doi.org/10.1109/TCDS.2020.2991789
Haller M, Case J, Crone NE, Chang EF, King-Stephens D, Laxer KD, Weber PB, Parvizi J, Knight RT, Shestyuk AY (2018) Persistent neuronal activity in human prefrontal cortex links perception and action. Nat Human Behav 2(1):80. https://doi.org/10.1038/s41562-017-0267-2
Herrnstein RJ (1961) Relative and absolute strength of response as a function of frequency of reinforcement. J Exp Anal Behav 4(3):267–272. https://doi.org/10.1901/jeab.1961.4-267
Histed MH, Pasupathy A, Miller EK (2009) Learning substrates in the primate prefrontal cortex and striatum: sustained activity related to successful actions. Neuron 63(2):244–253. https://doi.org/10.1016/j.neuron.2009.06.019
Hu SJ, Ko J, Weyand L, ElMaraghy HA, Lien TK, Koren Y, Bley H, Chryssolouris G, Nasr N, Shpitalni M (2011) Assembly system design and operations for product variety. CIRP Ann 60(2):715–733. https://doi.org/10.1016/j.cirp.2011.05.004
Huang CM, Mutlu B (2016) Anticipatory robot control for efficient human-robot collaboration. In: The eleventh ACM/IEEE international conference on human robot interaction, IEEE Press, pp 83–90, https://doi.org/10.1109/HRI.2016.7451737
Iigaya K, Ahmadian Y, Sugrue LP, Corrado GS, Loewenstein Y, Newsome WT, Fusi S (2019) Deviation from the matching law reflects an optimal strategy involving learning over multiple timescales. Nat Commun 10(1):1466. https://doi.org/10.1038/s41467-019-09388-3
Koene A, Remazeilles A, Prada M, Garzo A, Puerto M, Endo S, Wing AM (2014) Relative importance of spatial and temporal precision for user satisfaction in human-robot object handover interactions. In: Third International Symposium on New Frontiers in Human-Robot Interaction
Koulakov AA, Raghavachari S, Kepecs A, Lisman JE (2002) Model for a robust neural integrator. Nat Neurosci 5(8):775. https://doi.org/10.1038/nn893
Kozma R (2008) Intentional systems: review of neurodynamics, modeling, and robotics implementation. Phys Life Rev 5(1):1–21. https://doi.org/10.1016/j.plrev.2007.10.002
Krüger J, Lien TK, Verl A (2009) Cooperation of human and machines in assembly lines. CIRP Ann 58(2):628–646. https://doi.org/10.1016/j.cirp.2009.09.009
Laing CR, Troy WC, Gutkin B, Ermentrout GB (2002) Multiple bumps in a neuronal model of working memory. SIAM J Appl Math 63(1):62–97. https://doi.org/10.1137/S0036139901389495
Lau B, Glimcher PW (2005) Dynamic response-by-response models of matching behavior in rhesus monkeys. J Exp Anal Behav 84(3):555–579. https://doi.org/10.1901/jeab.2005.110-04
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553):436–444. https://doi.org/10.1038/nature14539
Lemaignan S, Warnier M, Sisbot EA, Clodic A, Alami R (2017) Artificial cognition for social human-robot interaction:an implementation. Artif Intell 247:45–69. https://doi.org/10.1016/j.artint.2016.07.002
Lin Y (2017) Toward intelligent human machine interactions. Mech Eng 139(06):S4–S8. https://doi.org/10.1115/1.2017-Jun-4
Lomp O, Richter M, Zibner SK, Schöner G (2016) Developing dynamic field theory architectures for embodied cognitive systems with cedar. Front Neurorobotics 10:14. https://doi.org/10.3389/fnbot.2016.00014
Louro L, Malheiro T, Guimarães P, Machado T, Monteiro S, Vaz Silva S, Erlhagen W, Bicho E (2019) Motion control for autonomous Tugger vehicles in dynamic factory floors shared with human operators. In: IEEE 45th Annual Conference of Industrial Electronics Society (IECON’2019), Lisbon, Portugal, October 14-17
Machado T, Malheiro T, Monteiro S, Erlhagen W, Bicho E (2019) Attractor dynamics approach to joint transportation by autonomous robots: theory, implementation and validation on the factory floor. Auton Robots 43(3):589–610. https://doi.org/10.1007/s10514-018-9729-2
Mayer MP, Schlick CM, Ewert D, Behnen D, Kuz S, Odenthal B, Kausch B (2011) Automation of robotic assembly processes on the basis of an architecture of human cognition. Product Eng 5(4):423–431. https://doi.org/10.1007/s11740-011-0316-z
Pardowitz M, Knoop S, Dillmann R, Zollner RD (2007) Incremental learning of tasks from user demonstrations, past experiences, and vocal comments. IEEE Trans Syst Man Cybern Part B Cybern 37(2):322–332. https://doi.org/10.1109/TSMCB.2006.886951
Pinheiro M, Bicho E, Erlhagen W (2010) A dynamic neural field architecture for a pro-active assistant robot. In: 2010 3rd IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics, IEEE, pp 777–784, https://doi.org/10.1109/BIOROB.2010.5627812
Rankin J, Avitabile D, Baladron J, Faye G, Lloyd DJ (2014) Continuation of localized coherent structures in nonlocal neural field equations. SIAM J Sci Comput 36(1):B70–B93. https://doi.org/10.1137/130918721
Remington ED, Egger SW, Narain D, Wang J, Jazayeri M (2018) A dynamical systems perspective on flexible motor timing. Trends Cognit Sci 22(10):938–952. https://doi.org/10.1016/j.tics.2018.07.010
Rhodes BJ, Bullock D, Verwey WB, Averbeck BB, Page MP (2004) Learning and production of movement sequences: behavioral, neurophysiological, and modeling perspectives. Human Movement Sci 23(5):699–746. https://doi.org/10.1016/j.humov.2004.10.008
Sakai Y, Okamoto H, Fukai T (2006) Computational algorithms and neuronal network models underlying decision processes. Neural Netw 19(8):1091–1105. https://doi.org/10.1016/j.neunet.2006.05.034
Schöner G (2019) The dynamics of neural populations capture the laws of the mind. Topics Cognit Sci. https://doi.org/10.1111/tops.12453
Seung HS, Lee DD, Reis BY, Tank DW (2000) Stability of the memory of eye position in a recurrent network of conductance-based model neurons. Neuron 26(1):259–271. https://doi.org/10.1016/s0896-6273(00)81155-1
Silva EC, Costa M, Araújo J, Machado D, Louro L, Erlhagen W, Bicho E (2015) Towards human-like bimanual movements in anthropomorphic robots: a nonlinear optimization approach. Appl Math Inf Sci Int J. https://doi.org/10.12785/amis/090210
Silva R, Louro L, Malheiro T, Erlhagen W, Bicho E (2016) Combining intention and emotional state inference in a dynamic neural field architecture for human-robot joint action. Adapt Behav 24(5):350–372. https://doi.org/10.1177/1059712316665451
Sousa E, Erlhagen W, Ferreira F, Bicho E (2015) Off-line simulation inspires insight: a neurodynamics approach to efficient robot task learning. Neural Netw 72:123–139. https://doi.org/10.1016/j.neunet.2015.09.002
Sugrue LP, Corrado GS, Newsome WT (2004) Matching behavior and the representation of value in the parietal cortex. Science 304(5678):1782–1787. https://doi.org/10.1126/science.1094765
Sünderhauf N, Brock O, Scheirer W, Hadsell R, Fox D, Leitner J, Upcroft B, Abbeel P, Burgard W, Milford M et al (2018) The limits and potentials of deep learning for robotics. Int J Robot Res 37(4–5):405–420. https://doi.org/10.1177/0278364918770733
Tsarouchi P, Makris S, Chryssolouris G (2016) Human-robot interaction review and challenges on task planning and programming. Int J Comput Integr Manuf 29(8):916–931. https://doi.org/10.1080/0951192X.2015.1130251
Wang J, Dou R, Muddada RR, Zhang W (2018a) Management of a holistic supply chain network for proactive resilience: theory and case study. Comput Indus Eng 125:668–677. https://doi.org/10.1016/j.cie.2017.12.021
Wang P, Liu H, Wang L, Gao RX (2018b) Deep learning-based human motion recognition for predictive context-aware human-robot collaboration. CIRP Ann 67(1):17–20. https://doi.org/10.1016/j.cirp.2018.04.066
Wei W, Song H, Li W, Shen P, Vasilakos A (2017) Gradient-driven parking navigation using a continuous information potential field based on wireless sensor network. Inf Sci 408:100–114. https://doi.org/10.1016/j.ins.2017.04.042
Wilcox R, Nikolaidis S, Shah J (2013) Optimization of temporal dynamics for adaptive human-robot interaction in assembly manufacturing. Robotics. https://doi.org/10.15607/RSS.2012.VIII.056
Wojtak W, Coombes S, Bicho E, Erlhagen W (2016) Combining spatial and parametric working memory in a dynamic neural field model. In: Villa A, Masulli P, Pons Rivero A (eds) Artificial Neural Networks and Machine Learning—ICANN 2016, Lecture Notes in Computer Science, Springer, vol 9886, pp 411–418, https://doi.org/10.1007/978-3-319-44778-0_48
Wojtak W, Ferreira F, Bicho E, Erlhagen W (2019) Neural field model for measuring and reproducing time intervals. In: Tetko IV, Kůrková V, Karpov P, Theis F (eds) Artificial Neural Networks and Machine Learning - ICANN 2019: Theoretical neural computation. Springer, Cham, pp 327–338. https://doi.org/10.1007/978-3-030-30487-4_26
Zunino A, Cavazza J, Volpi R, Morerio P, Cavallo A, Becchio C, Murino V (2020) Predicting intentions from motion: the subject-adversarial adaptation approach. Int J Comput Vis 128(1):220–239. https://doi.org/10.1007/s11263-019-01234-9
Acknowledgements
The work received financial support from FCT through the PhD fellowships PD/BD/128183/2016 and SFRH/BD/124912/2016, the project “Neurofield” (PTDC/MAT-APL/31393/2017) and the research centre CMAT within the project UID/MAT/00013/2020.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Initial conditions and parameters
Initial conditions and parameters
1.1 Initial conditions
1.1.1 Assembly task
For the model simulations, the initial conditions of the fields governed by the Amari dynamics, \(u_{per}\), \(u_{on}\), \(u_{wm}\) and \(u_d\) are defined by the inhibition parameter h. For the coupled two-field model the initial conditions are given by:
1.1.2 Value-based decision-making task
The initial condition of the decision field at the start of simulation trial n is given by:
The initial condition for the choice integration layer \((u_c,v_c)\) in the first trial and after each reset is given by
The initial condition for the success integration layer \((u_r,v_r)\) in the first trial and after each reset is given by
1.2 Model parameters
1.3 Numerical model simulations
Numerical simulations of the model were done in MATLAB using a forward Euler method with parameters given in Table 6. To compute the spatial convolution of w and f we employ a fast Fourier transform (FFT), using MATLAB’s in-built functions fft and ifft to perform the Fourier transform and the inverse Fourier transform, respectively.
Rights and permissions
About this article
Cite this article
Wojtak, W., Ferreira, F., Vicente, P. et al. A neural integrator model for planning and value-based decision making of a robotics assistant. Neural Comput & Applic 33, 3737–3756 (2021). https://doi.org/10.1007/s00521-020-05224-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-020-05224-8