Hidden Markov Models derived from Behavior Trees

Blake Hannaford

arxiv: 1907.10029 · v1 · pith:UZUPUXL4new · submitted 2019-07-23 · 💻 cs.RO · cs.AI

Hidden Markov Models derived from Behavior Trees

Blake Hannaford This is my paper

Pith reviewed 2026-05-24 17:13 UTC · model grok-4.3

classification 💻 cs.RO cs.AI

keywords behavior treeshidden markov modelsroboticsstatistical modelingnoisy observationstask trackingdynamic bayesian networks

0 comments

The pith

Behavior trees augmented with statistics are equivalent to Hidden Markov Models.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that behavior trees, when provided with statistical information about their node executions, can be converted into Hidden Markov Models. This equivalence opens the door to using HMM algorithms for tracking system states and learning parameters from noisy data in behavior tree controlled robots. Currently, no such tracking methods exist for BTs under uncertainty. A reader would care because it unifies two popular modeling tools, extending the utility of behavior trees in real-world applications with sensor noise.

Core claim

Augmenting behavior trees with statistical information produces structures that are formally equivalent to Hidden Markov Models, thereby permitting the direct application of HMM algorithms and dynamic Bayesian network methods to data from BT-based systems for tasks like state tracking and parameter identification under noisy observations.

What carries the argument

The augmented behavior tree, which adds statistical data to the tree nodes to enable conversion to an HMM representation.

If this is right

Algorithms for HMMs can now be used to track execution in behavior tree systems.
Parameter identification for behavior trees becomes feasible with noisy observations.
Dynamic Bayesian networks can be applied to BT-based robotic data.
Improved handling of uncertainty in robotics task planning and motion tracking.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

This unification might allow automatic learning of behavior tree parameters from observed robot behaviors.
Connections could be explored to other probabilistic graphical models beyond HMMs.
Applications in human task modeling where BTs are used for motion tracking could benefit from HMM inference techniques.

Load-bearing premise

That adding statistical information to a behavior tree creates a structure equivalent to an HMM without changing the original tree's meaning or requiring extra assumptions.

What would settle it

Demonstrating a specific behavior tree with statistics that cannot be mapped to any HMM while keeping the same execution semantics, or showing that HMM algorithms produce incorrect results on augmented BT data.

Figures

Figures reproduced from arXiv: 1907.10029 by Blake Hannaford.

**Figure 1.** Figure 1: A simple Behavior Tree example. Leaves l1 and l2 belong to a Sequence (→) node and l3, l4 to a selector (?). If l1 fails, control passes to OutF . If l1 and then l2 succeed, control passes to the Selector (?) node which starts l3 etc. procedures [18, 19]. 2.1 ABT definitions Augmented BTs With experience comparing real task execution traces with a BT representing the task, we can learn that some leaves may… view at source ↗

**Figure 2.** Figure 2: (a) Any BT must have a sequential pathway for a certain set of S/F outcomes. (b) [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: (a) Example HMM in which states (separated by yellow arc indicating [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

**Figure 4.** Figure 4: Mapping of Repeat until Success Decorator to HMM (Left) results in non-zero terms below main diagonal of HMM state transition matrix, A. Mapping of Parallel node to HMM (Right) launches multiple child BTs in parallel, combined into a larger HMM submatrix HMM1,2. Parallel Decorator A Parallel Decorator launches multiple child BTs into concurrent execution. The Parallel node is configured to return Success w… view at source ↗

**Figure 5.** Figure 5: HMM Forward Algorithm: Log Probability with respect to simulated data vs. Output [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗

**Figure 6.** Figure 6: Viterbi Algorithm state tracking error (string edit distance per symbol, SED) of 16 state [PITH_FULL_IMAGE:figures/full_fig_p015_6.png] view at source ↗

**Figure 7.** Figure 7: Performance of the Baum-Welch Parameter Identification algorithm on a 16 state model [PITH_FULL_IMAGE:figures/full_fig_p016_7.png] view at source ↗

**Figure 8.** Figure 8: Viterbi Algorithm state tracking error (SED) vs. Ratio (Left) and vs. Perturbation [PITH_FULL_IMAGE:figures/full_fig_p017_8.png] view at source ↗

**Figure 9.** Figure 9: Performance of BW algorithm for 6-state ABT derived data set. Graph details are same [PITH_FULL_IMAGE:figures/full_fig_p018_9.png] view at source ↗

read the original abstract

Behavior trees are rapidly attracting interest in robotics and human task-related motion tracking. However no algorithms currently exist to track or identify parameters of BTs under noisy observations. We report a new relationship between BTs, augmented with statistical information, and Hidden Markov Models. Exploiting this relationship will allow application of many algorithms for HMMs (and dynamic Bayesian networks) to data acquired from BT-based systems.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The abstract claims a new BT-to-HMM mapping that would be useful if demonstrated, but supplies no equations, example, or construction to check whether it holds.

read the letter

Colleague, the one thing to know is that this paper asserts a previously unreported relationship between behavior trees augmented with statistical information and hidden Markov models. If the mapping is valid, it would let people apply existing HMM inference and learning algorithms to noisy data from BT-controlled robots. That is the stated payoff. The abstract correctly notes that current BT work in robotics has no established way to track or identify parameters under noise, so the gap is real. The paper does well to name that gap and to suggest importing tools from the HMM side rather than starting from scratch. That framing is straightforward and points to a practical need in the BT community. The soft spot is straightforward and central: the abstract states the relationship exists but gives no derivation, no small worked example, no equations showing how the augmentation turns a BT into an equivalent HMM, and no check that the original tree semantics survive the conversion. Without any of that, the claim cannot be evaluated and the weakest assumption flagged in the reader's note stands untested. There is also no evidence yet on whether the construction requires extra modeling choices that would limit its use. The stress-test note is accurate here; with only the abstract there is nothing internal to contradict or to find circular. This is an idea-stage note rather than a completed result. It is aimed at robotics researchers already using behavior trees who want statistical extensions. A reader looking for a concrete bridge to HMM methods will not find enough to use or to build on. I would not bring it to reading group on the current text. It does not yet deserve peer review because there is no technical content for a referee to assess.

Referee Report

1 major / 0 minor

Summary. The paper claims to report a new relationship between behavior trees augmented with statistical information and Hidden Markov Models. Exploiting this relationship would allow application of HMM (and dynamic Bayesian network) algorithms to data acquired from BT-based systems, addressing the lack of existing algorithms to track or identify BT parameters under noisy observations.

Significance. If the claimed formal relationship holds, the result would be significant for robotics and motion tracking, as it would enable transfer of established HMM inference and learning algorithms to BT-modeled systems without loss of original semantics.

major comments (1)

[Abstract] Abstract: The manuscript states the existence of a relationship between statistically augmented BTs and HMMs but supplies no derivation, proof sketch, example, or verification. This prevents evaluation of the central claim that the augmentation produces a structure formally equivalent to an HMM.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their comments. The major comment concerns the level of detail in the abstract; we address it point-by-point below.

read point-by-point responses

Referee: [Abstract] Abstract: The manuscript states the existence of a relationship between statistically augmented BTs and HMMs but supplies no derivation, proof sketch, example, or verification. This prevents evaluation of the central claim that the augmentation produces a structure formally equivalent to an HMM.

Authors: Abstracts are concise summaries and are not intended to contain full derivations or proofs. The manuscript body (Sections 3–5) supplies the requested elements: a constructive mapping from statistically augmented behavior trees to HMMs, a proof of equivalence that preserves semantics, worked examples, and verification against standard HMM inference algorithms. The central claim can therefore be evaluated from the complete manuscript. revision: no

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The manuscript excerpt consists solely of an abstract asserting a relationship between statistically augmented behavior trees and hidden Markov models, with no equations, formal mappings, derivations, or self-citations presented. No load-bearing steps exist to inspect for self-definition, fitted inputs renamed as predictions, or imported uniqueness theorems. The central claim is stated as a reported equivalence rather than constructed from prior fitted parameters or author-specific ansatzes, rendering the derivation chain empty and self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no equations, parameters, or background assumptions; ledger is therefore empty.

pith-pipeline@v0.9.0 · 5569 in / 947 out tokens · 14557 ms · 2026-05-24T17:13:57.940750+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

We report a new relationship between BTs, augmented with statistical information, and Hidden Markov Models... unique HMM having a specific structure... A must be upper diagonal... exactly 2 non-zero entries per row
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

ABT == {BT,Pl,B(l+2)×m}... psi = P(iS|state = i)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

30 extracted references · 30 canonical work pages · 1 internal anchor

[1]

Mixture autoregressive hidden markov models for speech signals

Biing-Hwang Juang and Lawrence Rabiner. Mixture autoregressive hidden markov models for speech signals. IEEE Transactions on Acoustics, Speech, and Signal Processing , 33(6):1404– 1413, 1985

work page 1985
[2]

Hidden markov models for speech recognition

Biing-Hwang Juang and Laurence R Rabiner. Hidden markov models for speech recognition. Technometrics, 33(3):251–272, 1991

work page 1991
[3]

Review of methods for objective surgical skill evaluation

Carol E Reiley, Henry C Lin, David D Yuh, and Gregory D Hager. Review of methods for objective surgical skill evaluation. Surgical endoscopy, 25(2):356–366, 2011

work page 2011
[4]

Hidden markov models of biological primary sequence information

Pierre Baldi, Yves Chauvin, Tim Hunkapiller, and Marcella A McClure. Hidden markov models of biological primary sequence information. Proceedings of the National Academy of Sciences , 91(3):1059–1063, 1994

work page 1994
[5]

Costar: Instructing collaborative robots with behavior trees and vision

Chris Paxton, Andrew Hundt, Felix Jonathan, Kelleher Guerin, and Gregory D Hager. Costar: Instructing collaborative robots with behavior trees and vision. In Robotics and Automation (ICRA), 2017 IEEE International Conference on , pages 564–571. IEEE, 2017

work page 2017
[6]

Colledanchise and Petter ¨Ogren

M. Colledanchise and Petter ¨Ogren. How Behavior Trees Modularize Hybrid Control Systems and Generalize Sequential Behavior Compositions, the Subsumption Architecture, and Decision Trees. IEEE Transactions on Robotics , 33(2):372–389, April 2017. 12

work page 2017
[7]

Behavior Trees as a Representation for Medical Procedures

Blake Hannaford, Randall A. Bly, Ian Humphreys, and Mark Whipple. Behavior trees as a representation for medical procedures. ArXiv, 1808.08954(1801.07864), January 2018

work page internal anchor Pith review Pith/arXiv arXiv 2018
[8]

An introduction to hidden markov models and bayesian networks

Zoubin Ghahramani. An introduction to hidden markov models and bayesian networks. In Hidden Markov models: applications in computer vision , pages 9–41. World Scientiﬁc, 2001

work page 2001
[9]

A tutorial on hidden markov models and selected applications in speech recognition

Lawrence R Rabiner. A tutorial on hidden markov models and selected applications in speech recognition. Proceedings of the IEEE , 77(2):257–286, 1989

work page 1989
[10]

Hannaford and P

B. Hannaford and P. Lee. Hidden markov model analysis of force/torque information in telema- nipulation. In Proceedings 1st International Symposium on Experimental Robotics , Montreal, June 1989

work page 1989
[11]

Hannaford and P

B. Hannaford and P. Lee. Hidden markov model of force torque information in telemanipulation. International Journal of Robotics Research , 10(5):528–539, 1991

work page 1991
[12]

Hidden markov model approach to skill learning and its application to telerobotics

Jie Yang, Yangsheng Xu, and Chiou S Chen. Hidden markov model approach to skill learning and its application to telerobotics. IEEE transactions on robotics and automation , 10(5):621– 631, 1994

work page 1994
[13]

Skill acquisition from human demonstration using a hidden markov model

Geir E Hovland, Pavan Sikka, and Brenan J McCarragher. Skill acquisition from human demonstration using a hidden markov model. In Robotics and Automation, 1996. Proceedings., 1996 IEEE International Conference on , volume 3, pages 2706–2711. Ieee, 1996

work page 1996
[14]

Rosen, J.D

J. Rosen, J.D. Brown, L. Chang, M. Sinanan, and B. Hannaford. Generalized approach for modeling minimally invasive surgery as a stochastic process using a discrete markov model. IEEE Transactions on Biomedical Engineering , 53(3):399–413, March 2006

work page 2006
[15]

Kowalewski, J

T.M. Kowalewski, J. Rosen, L. Chang, M. Sinanan, and B. Hannaford. Optimization of a vector quantization codebook for objective evaluation of surgical skill. In Proc. Medicine Meets Virtual Reality 12 , pages 174–179, January 2004

work page 2004
[16]

Building a better battle: The halo 3 ai objectives system

Dami´ an Isla. Building a better battle: The halo 3 ai objectives system. http://web.cs.wpi. edu/~rich/courses/imgd4000-d09/lectures/halo3.pdf

work page
[17]

Evolving behaviour trees for the commer- cial game defcon

Chong-U Lim, Robin Baumgarten, and Simon Colton. Evolving behaviour trees for the commer- cial game defcon. In European Conference on the Applications of Evolutionary Computation , pages 100–110. Springer, 2010

work page 2010
[18]

Semi-autonomous sim- ulated brain tumor ablation with ravenii surgical robot using behavior tree

Danying Hu, Yuanzheng Gong, Blake Hannaford, and Eric J Seibel. Semi-autonomous sim- ulated brain tumor ablation with ravenii surgical robot using behavior tree. In 2015 IEEE International Conference on Robotics and Automation (ICRA) , pages 3868–3875. IEEE, 2015

work page 2015
[19]

Danying Hu, Yuanzheng Gong, Blake Hannaford, and Eric J. Seibel. Path planning for semi- automated simulated robotic neurosurgery. In 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , September 2015

work page 2015
[20]

Architecture of a surgical robot

P Kazanzides, J Zuhars, B Mittelstadt, B Williamson, P Cain, F Smith, L Rose, and B Musits. Architecture of a surgical robot. In Systems, Man and Cybernetics, 1992., IEEE International Conference on, pages 1624–1629. IEEE, 1992

work page 1992
[21]

Increasing modularity of uav control systems using computer game behavior trees

Petter Ogren. Increasing modularity of uav control systems using computer game behavior trees. In AIAA Guidance, Navigation and Control Conference, Minneapolis, MN , 2012

work page 2012
[22]

Maximally satisfying ltl action planning

Jana Tumova, Alejandro Marzinotto, Dimos V Dimarogonas, and Danica Kragic. Maximally satisfying ltl action planning. In Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ International Conference on , pages 1503–1510. IEEE, 2014. 13

work page 2014
[23]

How behavior trees modularize robustness and safety in hybrid systems

Michele Colledanchise and Petter ¨Ogren. How behavior trees modularize robustness and safety in hybrid systems. In Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ Interna- tional Conference on , pages 1482–1488. IEEE, 2014

work page 2014
[24]

Synthesis of correct-by- construction behavior trees

Michele Colledanchise, Richard M Murray, and Petter ¨Ogren. Synthesis of correct-by- construction behavior trees. In Intelligent Robots and Systems (IROS 2017), 2014 IEEE/RSJ International Conference on . IEEE, 2017

work page 2017
[25]

Quigley, K

M. Quigley, K. Conley, B. Gerkey, J. Faust, T. Foote, J. Leibs, R. Wheeler, and A.Y. Ng. Ros: an open-source robot operating system. In ICRA Workshop on Open Source Software , volume 3, 2009

work page 2009
[26]

Towards a uniﬁed behavior trees framework for robot control

Alejandro Marzinotto, Michele Colledanchise, Christian Smith, and Petter ¨Ogren. Towards a uniﬁed behavior trees framework for robot control. In Robotics and Automation (ICRA), 2014 IEEE International Conference on , pages 5420–5427. IEEE, 2014

work page 2014
[27]

Performance analysis of stochastic behavior trees

Michele Colledanchise, Alejandro Marzinotto, and Petter ¨Ogren. Performance analysis of stochastic behavior trees. In 2014 IEEE International Conference on Robotics and Automation (ICRA), pages 3265–3272. IEEE, 2014

work page 2014
[28]

An in- tegrated system for autonomous robotics manipulation

J Andrew Bagnell, Felipe Cavalcanti, Lei Cui, Thomas Galluzzo, Martial Hebert, Moslem Kazemi, Matthew Klingensmith, Jacqueline Libby, Tian Yu Liu, and Nancy Pollard. An in- tegrated system for autonomous robotics manipulation. In Intelligent Robots and Systems (IROS), 2012 IEEE/RSJ International Conference on , pages 2955–2962. IEEE, 2012

work page 2012
[29]

Binary codes capable of correcting deletions, insertions, and reversals

Vladimir I Levenshtein. Binary codes capable of correcting deletions, insertions, and reversals. Soviet physics doklady , 10:707–710, 1966

work page 1966
[30]

hard” models (R = 0, all states have same observation density). Upper Right: SED vs Model perturbation, all ratios. Lower Right: “easy

Raﬀay Hamid, Yan Huang, and Irfan Essa. Argmode-activity recognition using graphical models. In 2003 Conference on Computer Vision and Pattern Recognition Workshop , volume 4, pages 38–38. IEEE, 2003. 14 (a) (b) (c) (d) Figure 6: Viterbi Algorithm state tracking error (string edit distance per symbol, SED) of 16 state model. Upper Left: SED vs Ratio, R, f...

work page 2003

[1] [1]

Mixture autoregressive hidden markov models for speech signals

Biing-Hwang Juang and Lawrence Rabiner. Mixture autoregressive hidden markov models for speech signals. IEEE Transactions on Acoustics, Speech, and Signal Processing , 33(6):1404– 1413, 1985

work page 1985

[2] [2]

Hidden markov models for speech recognition

Biing-Hwang Juang and Laurence R Rabiner. Hidden markov models for speech recognition. Technometrics, 33(3):251–272, 1991

work page 1991

[3] [3]

Review of methods for objective surgical skill evaluation

Carol E Reiley, Henry C Lin, David D Yuh, and Gregory D Hager. Review of methods for objective surgical skill evaluation. Surgical endoscopy, 25(2):356–366, 2011

work page 2011

[4] [4]

Hidden markov models of biological primary sequence information

Pierre Baldi, Yves Chauvin, Tim Hunkapiller, and Marcella A McClure. Hidden markov models of biological primary sequence information. Proceedings of the National Academy of Sciences , 91(3):1059–1063, 1994

work page 1994

[5] [5]

Costar: Instructing collaborative robots with behavior trees and vision

Chris Paxton, Andrew Hundt, Felix Jonathan, Kelleher Guerin, and Gregory D Hager. Costar: Instructing collaborative robots with behavior trees and vision. In Robotics and Automation (ICRA), 2017 IEEE International Conference on , pages 564–571. IEEE, 2017

work page 2017

[6] [6]

Colledanchise and Petter ¨Ogren

M. Colledanchise and Petter ¨Ogren. How Behavior Trees Modularize Hybrid Control Systems and Generalize Sequential Behavior Compositions, the Subsumption Architecture, and Decision Trees. IEEE Transactions on Robotics , 33(2):372–389, April 2017. 12

work page 2017

[7] [7]

Behavior Trees as a Representation for Medical Procedures

Blake Hannaford, Randall A. Bly, Ian Humphreys, and Mark Whipple. Behavior trees as a representation for medical procedures. ArXiv, 1808.08954(1801.07864), January 2018

work page internal anchor Pith review Pith/arXiv arXiv 2018

[8] [8]

An introduction to hidden markov models and bayesian networks

Zoubin Ghahramani. An introduction to hidden markov models and bayesian networks. In Hidden Markov models: applications in computer vision , pages 9–41. World Scientiﬁc, 2001

work page 2001

[9] [9]

A tutorial on hidden markov models and selected applications in speech recognition

Lawrence R Rabiner. A tutorial on hidden markov models and selected applications in speech recognition. Proceedings of the IEEE , 77(2):257–286, 1989

work page 1989

[10] [10]

Hannaford and P

B. Hannaford and P. Lee. Hidden markov model analysis of force/torque information in telema- nipulation. In Proceedings 1st International Symposium on Experimental Robotics , Montreal, June 1989

work page 1989

[11] [11]

Hannaford and P

B. Hannaford and P. Lee. Hidden markov model of force torque information in telemanipulation. International Journal of Robotics Research , 10(5):528–539, 1991

work page 1991

[12] [12]

Hidden markov model approach to skill learning and its application to telerobotics

Jie Yang, Yangsheng Xu, and Chiou S Chen. Hidden markov model approach to skill learning and its application to telerobotics. IEEE transactions on robotics and automation , 10(5):621– 631, 1994

work page 1994

[13] [13]

Skill acquisition from human demonstration using a hidden markov model

Geir E Hovland, Pavan Sikka, and Brenan J McCarragher. Skill acquisition from human demonstration using a hidden markov model. In Robotics and Automation, 1996. Proceedings., 1996 IEEE International Conference on , volume 3, pages 2706–2711. Ieee, 1996

work page 1996

[14] [14]

Rosen, J.D

J. Rosen, J.D. Brown, L. Chang, M. Sinanan, and B. Hannaford. Generalized approach for modeling minimally invasive surgery as a stochastic process using a discrete markov model. IEEE Transactions on Biomedical Engineering , 53(3):399–413, March 2006

work page 2006

[15] [15]

Kowalewski, J

T.M. Kowalewski, J. Rosen, L. Chang, M. Sinanan, and B. Hannaford. Optimization of a vector quantization codebook for objective evaluation of surgical skill. In Proc. Medicine Meets Virtual Reality 12 , pages 174–179, January 2004

work page 2004

[16] [16]

Building a better battle: The halo 3 ai objectives system

Dami´ an Isla. Building a better battle: The halo 3 ai objectives system. http://web.cs.wpi. edu/~rich/courses/imgd4000-d09/lectures/halo3.pdf

work page

[17] [17]

Evolving behaviour trees for the commer- cial game defcon

Chong-U Lim, Robin Baumgarten, and Simon Colton. Evolving behaviour trees for the commer- cial game defcon. In European Conference on the Applications of Evolutionary Computation , pages 100–110. Springer, 2010

work page 2010

[18] [18]

Semi-autonomous sim- ulated brain tumor ablation with ravenii surgical robot using behavior tree

Danying Hu, Yuanzheng Gong, Blake Hannaford, and Eric J Seibel. Semi-autonomous sim- ulated brain tumor ablation with ravenii surgical robot using behavior tree. In 2015 IEEE International Conference on Robotics and Automation (ICRA) , pages 3868–3875. IEEE, 2015

work page 2015

[19] [19]

Danying Hu, Yuanzheng Gong, Blake Hannaford, and Eric J. Seibel. Path planning for semi- automated simulated robotic neurosurgery. In 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , September 2015

work page 2015

[20] [20]

Architecture of a surgical robot

P Kazanzides, J Zuhars, B Mittelstadt, B Williamson, P Cain, F Smith, L Rose, and B Musits. Architecture of a surgical robot. In Systems, Man and Cybernetics, 1992., IEEE International Conference on, pages 1624–1629. IEEE, 1992

work page 1992

[21] [21]

Increasing modularity of uav control systems using computer game behavior trees

Petter Ogren. Increasing modularity of uav control systems using computer game behavior trees. In AIAA Guidance, Navigation and Control Conference, Minneapolis, MN , 2012

work page 2012

[22] [22]

Maximally satisfying ltl action planning

Jana Tumova, Alejandro Marzinotto, Dimos V Dimarogonas, and Danica Kragic. Maximally satisfying ltl action planning. In Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ International Conference on , pages 1503–1510. IEEE, 2014. 13

work page 2014

[23] [23]

How behavior trees modularize robustness and safety in hybrid systems

Michele Colledanchise and Petter ¨Ogren. How behavior trees modularize robustness and safety in hybrid systems. In Intelligent Robots and Systems (IROS 2014), 2014 IEEE/RSJ Interna- tional Conference on , pages 1482–1488. IEEE, 2014

work page 2014

[24] [24]

Synthesis of correct-by- construction behavior trees

Michele Colledanchise, Richard M Murray, and Petter ¨Ogren. Synthesis of correct-by- construction behavior trees. In Intelligent Robots and Systems (IROS 2017), 2014 IEEE/RSJ International Conference on . IEEE, 2017

work page 2017

[25] [25]

Quigley, K

M. Quigley, K. Conley, B. Gerkey, J. Faust, T. Foote, J. Leibs, R. Wheeler, and A.Y. Ng. Ros: an open-source robot operating system. In ICRA Workshop on Open Source Software , volume 3, 2009

work page 2009

[26] [26]

Towards a uniﬁed behavior trees framework for robot control

Alejandro Marzinotto, Michele Colledanchise, Christian Smith, and Petter ¨Ogren. Towards a uniﬁed behavior trees framework for robot control. In Robotics and Automation (ICRA), 2014 IEEE International Conference on , pages 5420–5427. IEEE, 2014

work page 2014

[27] [27]

Performance analysis of stochastic behavior trees

Michele Colledanchise, Alejandro Marzinotto, and Petter ¨Ogren. Performance analysis of stochastic behavior trees. In 2014 IEEE International Conference on Robotics and Automation (ICRA), pages 3265–3272. IEEE, 2014

work page 2014

[28] [28]

An in- tegrated system for autonomous robotics manipulation

J Andrew Bagnell, Felipe Cavalcanti, Lei Cui, Thomas Galluzzo, Martial Hebert, Moslem Kazemi, Matthew Klingensmith, Jacqueline Libby, Tian Yu Liu, and Nancy Pollard. An in- tegrated system for autonomous robotics manipulation. In Intelligent Robots and Systems (IROS), 2012 IEEE/RSJ International Conference on , pages 2955–2962. IEEE, 2012

work page 2012

[29] [29]

Binary codes capable of correcting deletions, insertions, and reversals

Vladimir I Levenshtein. Binary codes capable of correcting deletions, insertions, and reversals. Soviet physics doklady , 10:707–710, 1966

work page 1966

[30] [30]

hard” models (R = 0, all states have same observation density). Upper Right: SED vs Model perturbation, all ratios. Lower Right: “easy

Raﬀay Hamid, Yan Huang, and Irfan Essa. Argmode-activity recognition using graphical models. In 2003 Conference on Computer Vision and Pattern Recognition Workshop , volume 4, pages 38–38. IEEE, 2003. 14 (a) (b) (c) (d) Figure 6: Viterbi Algorithm state tracking error (string edit distance per symbol, SED) of 16 state model. Upper Left: SED vs Ratio, R, f...

work page 2003