Pith. sign in

REVIEW 4 major objections 5 minor 35 references

Hanging Around: Cognitive Inspired Reasoning for Reactive Robotics

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A robot learns to recognize handles by watching mugs hang from hooks

desk verdict Good idea, missing demonstration: the paper defines a functional-part discovery loop but never measures it. read the letter →

arxiv 2507.20832 v1 pith:AJEJUJFJ submitted 2025-07-28 cs.RO

Mihai Pomarlan , Stefano De Giorgis , Rachel Ringe , Maria M. Hedblom , Nikolaos Tsiogkas This is my paper · ORCID
classification cs.RO
keywords neurosymbolicarchitectureimageschemasfunctionalpartssupportrelationsreactiveroboticsconceptdiscoveryperceptiongroundingontologyexpansion
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper builds a robot agent that starts with no concept of a handle and, by observing mugs supported by hooks, invents the concept of a functional part: the part of an object that does the supporting work. The agent combines a neural vision module with an ontological theory of Support, so that perception queries, belief updates, and planning all flow through the same symbolic description. The central claim is that once the symbolic theory identifies the contact region in an observed support situation, the agent can collect those regions as training data, retrain its detector to recognize the part even outside support scenes, and use that part to plan hanging and unhanging actions. The significance is a concrete route out of the fixed-ontology trap for reactive robots: new perceptual categories arise from sensorimotor observation and become usable in planning.

What carries the argument

The load-bearing mechanism is the image-schematic theory of Support coupled to a query-driven perception loop. Support is formalized using the Description-and-Situation pattern, with necessary roles supportee and supporter; axiom 16 turns observable symptoms, namely contact, supporter below, and supportee not moving down, into belief in a Support situation, while axiom 15 turns a believed Support description into perception queries for contact and movement relative to the floor. Contact masks returned by perception become the grounding data: axiom 17 defines MugSuppByHook as the part of a Mug that participates as supportee in a Support situation with a Hook, and those masks are stored as training examples. The same symbolic theory supplies constraints for a pose solver, so the discovered part simultaneously names a new perception class and focuses the planning search.

What would settle it

In the simulated world, take a mug in a support situation, extract the contact mask defined by axiom 16, retrain the detector on those masks, and then run the detector on scenes where the mug is not hanging to check that the detected MugSuppByHook region still coincides with the handle. If the region jumps between retraining runs or mismatches the handle, the claim that the network recognizes a functional part outside support situations fails.

Watch

Extended reading notes

Core claim

The central claim is that an agent with no concept of a handle can acquire one by observing supported objects hanging from a hook, and can then use that concept in planning. The symbolic theory of Support states that a support situation exists when a supportee is in contact with a supporter, the supporter is below, and the supportee does not move down (Eq. 16). Applying that theory to perception outputs yields contact masks, and the agent defines a new class, MugSuppByHook, for the part of a mug that participates as supportee in such a situation (Eq. 17). The paper argues that the retrained neural network can recognize MugSuppByHook even outside a supported-by-hook scene, and that solving pose constraints with this part rather than the whole mug makes a satisfying pose virtually guarantee that the mug is actually supported by the hook.

Load-bearing premise

The whole loop depends on the vision module's contact masks and segmentations being accurate enough that the symbolic Support theory always picks out the same reliable physical part, and the paper reports no accuracy measurements for this.

Editorial extensions

If this is right

  • Because the agent creates its own labels from contact masks, it can expand its perception without human-annotated data for the new part.
  • The new class keeps its functional meaning: the network recognizes the part that can be a supportee in a hook support, not just a visual pattern.
  • Planning for support and support-destruction becomes tractable: searching over the handle part instead of the whole mug removes unstable poses that touch the hook without hanging.
  • The same mechanism can in principle be replayed for other image schemas, so the agent's ontology is open-ended rather than fixed at design time.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension the paper leaves implicit is applying the same bootstrapping to other image schemas, such as Containment or Blockage, and measuring how much symbolic theory is needed to label data for each new functional part.
  • If the approach is ported to physical robots, the unmeasured quantity that will decide success is the stability of automatically selected contact masks; noisy real-world masks could make the new class drift between retraining runs.
  • The paper's 'virtually guaranteed' planning claim could be quantified as an unstable-pose rate: compare the fraction of constraint-solver solutions that fail to support when using the detected part versus the whole mug.
  • The architecture suggests a division of labor in which symbolic theories are deliberately small and only need to be accurate enough to label training data, not to model the world completely.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes a neurosymbolic architecture for reactive robotics in which a first-order-logic theory of the Support image schema (Section 4) drives perception queries, diagnoses support situations from contact, relative-below, and non-downward-motion predicates, and defines a new functional part class MugSuppByHook (Eq. 17). The agent is claimed to use contact masks from support situations as automatic training labels, retrain a YOLOv8 detector to recognize this functional part in novel, non-support contexts, and then use the detected part to constrain a geometric constraint solver so that planning a mug-on-hook support becomes feasible. The abstract states that the approach is demonstrated in a simulated world.

Significance. If the claimed learning loop worked as described, the paper would make a useful contribution to neurosymbolic robotics and concept discovery: the idea of bootstrapping a perceptual class from an image-schematic theory and using it to focus planning is genuinely interesting, and the FOL theory in Section 4 is a coherent naive formalization of Support. The authors also provide a public source-code link. However, the manuscript reports no experiments, no quantitative metrics, and no comparisons; the central claim is an unverified proposal rather than a demonstrated result. Moreover, the auto-labeling procedure is self-referential to a degree that makes the claimed generalization unmeasurable without an independent test set. As submitted, the significance of the contribution cannot be assessed.

major comments (4)
  1. [Abstract; Section 5] The paper's central claim—that the agent teaches itself to recognize a functional part and uses it to make hanging 'virtually guaranteed'—is not backed by any experimental results. The abstract announces a demonstration in a simulated world, but the manuscript contains no simulator description, no success rates for the planning loop, no precision/recall or accuracy for the retrained YOLO detector, and no comparison against a baseline. Without such measurements, the claimed 'demonstration' is a proposed mechanism, not a supported empirical result.
  2. [Eq. (17), Section 5] The learning loop is self-referential: MugSuppByHook is defined in Eq. (17) using DSupp and the perception predicates contact, below, and not moving down; the training labels are contact masks attached to situations selected by the same diagnostic rules; and success is judged by establishing a DSupp situation. There is no independent ground truth for what a handle is. Consequently, any systematic bias in YOLO segmentation, optical flow, or contact-region extraction is inherited by the new class, and the generalization claim ('recognize a MugSuppByHook even outside of a supported by Hook situation') requires a test set with manually annotated functional parts, which is not provided.
  3. [Eq. (16)] The diagnostic rule for DSupp is too weak: contact plus below plus not moving downward is also satisfied by an object resting on a table or shelf, so the auto-labeled data can mix hanging support with other support configurations. In addition, the consequent of Eq. (16) writes DSupp(e), applying the situation-description predicate to an object, whereas Eq. (15) applies DSupp(s) to a situation; as written the axiom is type-incoherent and cannot be used by the reasoner without correction.
  4. [Section 5, 'virtually guaranteed'] The statement that solving the constraint for the detected functional part makes it 'virtually guaranteed that if a pose satisfying the constraint is found, then the mug is in fact supported by the hook' is not justified. Contact plus 'mug above' plus non-overlap is not a sufficient condition for static stability (friction, center of mass, and force balance are not modeled), and the paper gives no probabilistic analysis or simulation data to support the guarantee. The claim should be removed or replaced by a measured success rate.
minor comments (5)
  1. [Eq. (7)] Axiom 7 is missing the conjunction before ¬exrt(o, f2); the displayed formula is not syntactically well-formed.
  2. [Section 1] There are several typos, e.g., 'Endorser of the ... claim' should be 'Endorsers' and 'Aspect such as' should be 'Aspects such as'; the text would benefit from copyediting.
  3. [Section 4] The paper states that some domain and range axioms are 'filled in by the reader'; for a formal theory this should be made explicit in an appendix or supplementary material, since the axioms as given are not a complete theory.
  4. [Figure 5, Eq. (17)] The symbol for the functional part is written both as 'MugSuppByHook' and 'MugSuppbyHook'; please use one consistent name.
  5. [Section 3.2] The contact mask is described informally as 'points near where this contact occurs'; the computation of 'near' and the mask extraction should be specified, preferably with parameters, to make the pipeline reproducible.

Circularity Check

2 steps flagged · score 4.0 of 10

Partially self-referential learning loop: the MugSuppByHook concept (Eq. 17), its auto-generated training labels, and the planning success criterion all reduce to the same Support diagnosis (Eq. 16: contact, below, not moving down), so the 'discovered' part re-detects the theory's own filter; the Support axioms themselves are nonetheless independent content, so circularity is only partial.

  1. self definitional [Section 5, Eq. 17; Section 4, Eq. 16]
    "∀x :MugSuppbyHook(x) ↔ (∃c, s, m, h : Con(c) ∧ DSupp(s) ∧ Hook (h) ∧ M ug(m) ∧ hasP rt(m, x) ∧ suppee(s, m) ∧ supper(s, h) ∧ hasP rtcp(c, x) ∧ hasP rtcp(c, h) ∧ below(h, x)) (17) ... Treating the definition of MugSuppByHook as a new concept allows the agent to collect images and contact masks that are observations of its instances, and retrain the neural network responsible for object detection."

    The new concept is defined directly in terms of the symbolic Support theory (DSupp, Con, below), and Eq. 16 is the diagnostic rule that fires on contact + below + not moving down. The training labels are the contact masks attached to exactly these DSupp situations, so by construction the 'discovered' functional part is the contact region the theory already selects: discovery, definition, and labels all reduce to Eq. 16's filter. The claimed payoff, 'Crucially, the network can recognize a MugSuppByHook even outside of a supported by Hook situation,' is an asserted generalization from this self-generated label set, with no reported precision/recall and no independent ground truth; any bias in YOLO segmentation, optical flow, or contact masks is baked into both the concept and its labels.

  2. fitted input called prediction [Section 5, final paragraph]
    "Using the part of a mug labeled as MugSuppByHook, instead of the whole mug, as the entity for which to solve constraints reduces the search space, and also makes it virtually guaranteed that if a pose satisfying the constraint is found, then the mug is in fact supported by the hook."

    The constraint solver is given contact plus mug-above-hook, which are the geometric conjuncts of Eq. 16's antecedent. 'In fact supported' is the full Eq. 16 diagnosis, which additionally requires ¬movDir(e,down), a condition the solver does not enforce. The 'virtually guaranteed' bridge transfers that missing condition from the training data, where it held by construction in every DSupp-labeled example. Since the success criterion is the same predicate family that generated the labels, the loop is self-confirming: a detector fitted to the theory's own diagnosis is used to vouch for the theory's own diagnosis, with no physics-based verification and no planning success rates reported.

full rationale

The derivation is not fully circular, which keeps the score moderate. Axioms 1-14 state an independent naive theory of Support (an object is supported iff it does not fall), grounded in external image-schematic literature; no numerical parameter is fitted to force the result, no uniqueness theorem is imported, and no load-bearing self-citation chain exists, since the gestalt-activation premise is attributed to [27,28] and self-citations such as [26] are contextual. However, the central discovery loop is self-referential at three connected points. First, Eq. 17 defines MugSuppByHook out of DSupp, Con, and below, the very predicates of Eq. 16 that generate the training labels (contact masks attached to DSupp situations), so by construction the discovered part is exactly what the theory already selects; the claim that the retrained network recognizes it outside support situations is asserted without measurements or independent ground truth. Second, the Section 5 'virtually guaranteed' claim fills the one condition the solver does not enforce, ¬movDir(e,down) from Eq. 16, by transferring it from the training set where it held by construction in every labeled DSupp example; success is thereby judged by the same predicate that produced the labels. Third, the paper's own Section 6 limitation, 'it is necessary to annotate functional parts on frames where they do not yet perform the function,' confirms that current annotations are made on frames where the part performs its function, so the claimed recognition outside supported situations is future work rather than a demonstrated result. These are partial circularity and a verification gap rather than an equation-forcing reduction, hence score 4; the lack of any experimental validation of the loop is better treated as a correctness risk than as additional circularity.

Assumptions & free parameters 0 free parameters · 5 assumptions · 1 invented entities

No numerical free parameters are reported. The theory rests on a set of domain assumptions about forces, contact, and support (Section 4), plus an assumed-reliable perception pipeline (Section 3.2). The one invented concept, MugSuppByHook, is generated internally and lacks independent validation.

assumptions (5)
  • domain assumption The floor exerts gravity, a downward force, on all objects (Eq. 3).
    Section 4. This idealization lets the theory distinguish support from other contact relations; it is a modeling choice, not an empirical result.
  • domain assumption Every contact between two objects implies mutual forces (Eq. 5).
    Section 4. Needed so that Support can be linked to contact queries.
  • domain assumption A typical object that does not move in the direction of a force must have an opposite force acting on it (Eq. 7).
    Section 4. This closes the force balance and lets the reasoner infer the existence of a supporter.
  • domain assumption A DSupp description can be diagnosed from contact, below, and not moving down (Eq. 16).
    Section 4. This symptom-based rule is what turns raw perception into training labels for the new functional part class; its accuracy is unverified.
  • domain assumption YOLOv8 segmentations, optical flow, and contact masks provide reliable qualitative descriptions.
    Section 3.2. The entire learning loop depends on these annotations, but no segmentation or detection accuracy is reported.
invented entities (1)
  • MugSuppByHook functional part class
    purpose: A concept for the part of a mug that plays the supportee role when the mug is supported by a hook; used to focus perception and planning.
    Defined by Eq. 17 from the agent's own support theory and trained from its own contact masks. No external benchmark confirms the class corresponds to a stable physical category.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hanging Around: Cognitive Inspired Reasoning for Reactive Robotics." pith.science (2026). https://pith.science/paper/AJEJUJFJ

@misc{pith2026250720832,
  author       = {Pith},
  title        = {Pith review of: Hanging Around: Cognitive Inspired Reasoning for Reactive Robotics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AJEJUJFJ}},
  note         = {Machine review of arXiv:2507.20832}
}
read the original abstract

Situationally-aware artificial agents operating with competence in natural environments face several challenges: spatial awareness, object affordance detection, dynamic changes and unpredictability. A critical challenge is the agent's ability to identify and monitor environmental elements pertinent to its objectives. Our research introduces a neurosymbolic modular architecture for reactive robotics. Our system combines a neural component performing object recognition over the environment and image processing techniques such as optical flow, with symbolic representation and reasoning. The reasoning system is grounded in the embodied cognition paradigm, via integrating image schematic knowledge in an ontological structure. The ontology is operatively used to create queries for the perception system, decide on actions, and infer entities' capabilities derived from perceptual data. The combination of reasoning and image processing allows the agent to focus its perception for normal operation as well as discover new concepts for parts of objects involved in particular interactions. The discovered concepts allow the robot to autonomously acquire training data and adjust its subsymbolic perception to recognize the parts, as well as making planning for more complex tasks feasible by focusing search on those relevant object parts. We demonstrate our approach in a simulated world, in which an agent learns to recognize parts of objects involved in support relations. While the agent has no concept of handle initially, by observing examples of supported objects hanging from a hook it learns to recognize the parts involved in establishing support and becomes able to plan the establishment/destruction of the support relation. This underscores the agent's capability to expand its knowledge through observation in a systematic way, and illustrates the potential of combining deep reasoning [...].

Figures

Figures reproduced from arXiv: 2507.20832 by the authors.

Figure 1
Figure 1. Architectural overview of the agent. 3.1. Towards Engagement with the World A naive understanding of perception would be that, modulo errors that in princi￾ple can be eliminated, it constructs a truthful picture of the world out of facts in￾dependent of contextual factors such as the goals of the perceiver. Following Hei￾degger, AI critic Hubert Dreyfus argued against this view and that it is responsi￾ble for the st… view at source ↗
Figure 2
Figure 2. Overview of the perception module. As shown in [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Left: a 3rd person view of the turtlebot in the scene. Middle: segmentation masks from YOLO. Right: optical flow points (purple) and contact masks (yellow) superimposed on the robot’s RGB image. The main perception output is a set of qualitative descriptions expressed as triples of forms (pso), (−pso) where p can be contacts, approaches, departs, stillness, and −p can be −contact (objects not in contact), and a set … view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Overview of the reasoning components. The main constituent of the robot’s belief is a set of persistent (image) schemas, i.e. assertions about relationships between objects such as Contact, Support, as well as assertions about the robot’s “goals”. Reification steps are…
Figure 5
Figure 5. Figure 5: Functional parts. Left: a frame stored for training, with annotations of functional parts. Right: frame where the newly trained network is used to recognize functional parts. The MugSuppByHook object is useful when the agent is given a goal to sup￾port the mug from a h…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

  1. [1]

    Sora’s surreal physics; 2024

    Marcus G. Sora’s surreal physics; 2024. Available from: https://garymarcus.substack. com/p/soras-surreal-physics

    work page 2024

  2. [2]

    Mind children: The future of robot and human intelligence

    Moravec H. Mind children: The future of robot and human intelligence. Harvard University Press; 1988

    work page 1988

  3. [3]

    Understanding the semantic web through descriptions and situations

    Gangemi A, Mika P. Understanding the semantic web through descriptions and situations. In: OTM Confederated International Conferences” On the Move to Meaningful Internet Systems”. Springer; 2003. p. 689-706

    work page 2003

  4. [4]

    Norms and plans as unification criteria for social collectives

    Gangemi A. Norms and plans as unification criteria for social collectives. Autonomous Agents and Multi-Agent Systems. 2008;17:70-112

    work page 2008

  5. [5]

    The Foundations of Mind: Origins of Conceptual Thought: Origins of Con- ceptual Though

    Mandler JM. The Foundations of Mind: Origins of Conceptual Thought: Origins of Con- ceptual Though. New York: Oxford University Press; 2004

    work page 2004

  6. [6]

    The Body in the Mind Metaphors

    Johnson M. The Body in the Mind Metaphors. University of Chicago Press; 1987

    work page 1987

  7. [7]

    The Invariance Hypothesis: is abstract reason based on image-schemas? Cognitive Linguistics

    LAKOFF G. The Invariance Hypothesis: is abstract reason based on image-schemas? Cognitive Linguistics. 1990;1(1):39-74. Available from: https://doi.org/10.1515/cogl. 1990.1.1.39

    work page doi:10.1515/cogl 1990

  8. [8]

    Where mathematics comes from

    Lakoff G, N´ u˜ nez R. Where mathematics comes from. vol. 6. New York: Basic Books; 2000

    work page 2000

Show all 35 references
  1. [9]

    An image-schematic account of spatial categories

    Kuhn W. An image-schematic account of spatial categories. In: International Conference on Spatial Information Theory. Springer; 2007. p. 152-68

    2007

  2. [10]

    Image schema combinations and complex events

    Hedblom MM, Kutz O, Pe˜ naloza R, Guizzardi G. Image schema combinations and complex events. KI-K¨ unstliche Intelligenz. 2019;33:279-91

    2019

  3. [11]

    An image schema language

    St Amant R, Morrison CT, Chang YH, Cohen PR, Beal C. An image schema language. In: International Conference on Cognitive Modeling (ICCM); 2006. p. 292-7

    2006

  4. [12]

    A framework for representing knowledge

    Minsky M. A framework for representing knowledge. MIT, Cambridge; 1974

    1974

  5. [13]

    Frame semantics

    Fillmore CJ. Frame semantics. In: Linguistics in the Morning Calm. Seoul: Hanshin; 1982. p. 111-38

    1982

  6. [14]

    F–a model of events based on the foundational ontology dolce+ DnS ultralight

    Scherp A, Franz T, Saathoff C, Staab S. F–a model of events based on the foundational ontology dolce+ DnS ultralight. In: Proceedings of the fifth international conference on Knowledge capture; 2009. p. 137-44

    2009

  7. [15]

    Formal ontology in information systems: proceedings of the Fifth International Conference (FOIS 2008)

    Eschenbach C, Gruninger M. Formal ontology in information systems: proceedings of the Fifth International Conference (FOIS 2008). vol. 183. IOS Press; 2008

    2008

  8. [16]

    Deep understanding of everyday activity commands for household robots

    H¨ offner S, Porzel R, Hedblom MM, Pomarlan M, Cangalovic VS, Pfau J, et al. Deep understanding of everyday activity commands for household robots. Semantic Web. 2022;13(5):895-909

    2022

  9. [17]

    A challenge for historical research: making data F AIR using a collaborative ontology management environment (OntoME)

    Beretta F. A challenge for historical research: making data F AIR using a collaborative ontology management environment (OntoME). Semantic Web. 2021;12(2):279-94

    2021

  10. [18]

    A spatial logic based on regions and connection

    Randell DA, Cui Z, Cohn AG. A spatial logic based on regions and connection. KR. 1992;92:165-76

    1992

  11. [19]

    A qualitative trajectory calculus as a basis for representing moving objects in geographical information systems

    Van de Weghe N, Cohn A, De Tre G, De Maeyer P. A qualitative trajectory calculus as a basis for representing moving objects in geographical information systems. Control and cybernetics. 2006;35(1):97-119

    2006

  12. [20]

    Temporal Logic and State Systems

    Kr¨ oger F, Merz S. Temporal Logic and State Systems. Texts in Theoretical Computer Science. An EATCS Series. Springer; 2008. Available from: https://doi.org/10.1007/ 978-3-540-68635-4

    2008

  13. [21]

    Image Schemas and Concept Invention: Cognitive, Logical, and Linguistic Investigations

    Hedblom MM. Image Schemas and Concept Invention: Cognitive, Logical, and Linguistic Investigations. Cognitive Technologies. Springer Computer Science; 2020

    2020

  14. [22]

    Between Contact and Support: Introducing a logic for image schemas and directed movement

    Hedblom MM, Kutz O, Mossakowski T, Neuhaus F. Between Contact and Support: Introducing a logic for image schemas and directed movement. In: Esposito F, Basili R, Ferilli S, Lisi F A, editors. AI*IA 2017: Advances in Artificial Intelligence; 2017. p. 256-68

    2017

  15. [23]

    Dynamic Action Selection Using Image Schema-Based Reasoning for Robots

    Hedblom MM, Pomarlan M, Porzel R, Malaka R, Beetz M. Dynamic Action Selection Using Image Schema-Based Reasoning for Robots. In: The 7th Joint Ontology Workshops (JOWO). Bolzano, Italy; 2021

    2021

  16. [24]

    Cutting Events: Towards Autonomous Plan Adaption by Robotic Agents through Image-Schematic Event Segmentation

    Dhanabalachandran K, Hassouna V, Hedblom MM, K¨ uempel M, Leusmann N, Beetz M. Cutting Events: Towards Autonomous Plan Adaption by Robotic Agents through Image-Schematic Event Segmentation. In: Proceedings of the 11th on Knowledge Capture Conference; 2021. p. 25-32

    2021

  17. [25]

    Imageschemanet: Formalizing embodied common- sense knowledge providing an imageschematic layer to framester

    De Giorgis S, Gangemi A, Gromann D. Imageschemanet: Formalizing embodied common- sense knowledge providing an imageschematic layer to framester. Semantic Web Journal, forthcoming. 2022

    2022

  18. [26]

    Thinking in front of the box: Towards intelligent robotic action selection for navigation in complex environments using image-schematic reasoning

    Pomarlan M, De Giorgis S, Hedblom M, Diab M, Tsiogkas N. Thinking in front of the box: Towards intelligent robotic action selection for navigation in complex environments using image-schematic reasoning. In: The 8th Joint Ontology Workshops (JOWO). J¨ onk¨ oping, Sweden; 2022

    2022

  19. [27]

    Gestalt psychology in Italy

    Verstegen I. Gestalt psychology in Italy. Journal of the History of the Behavioral Sciences. 2000;36(1):31-42

    2000

  20. [28]

    Metaphors we live by

    Lakoff G, Johnson M. Metaphors we live by. University of Chicago press; 1980

    1980

  21. [29]

    Cognitive robotics

    Levesque H, Lakemeyer G. Cognitive robotics. Foundations of artificial intelligence. 2008;3:869-86

    2008

  22. [30]

    Learning relational affordance models for robots in multi-object manipulation tasks

    Moldovan B, Moreno P, Van Otterlo M, Santos-Victor J, De Raedt L. Learning relational affordance models for robots in multi-object manipulation tasks. In: 2012 ieee international conference on robotics and automation. IEEE; 2012. p. 4373-8

    2012

  23. [31]

    Embodied cognitive robotics and the learning of sensorimotor schemes

    Lara B, Astorga D, Mendoza-Bock E, Pardo M, Escobar E, Ciria A. Embodied cognitive robotics and the learning of sensorimotor schemes. Adaptive Behavior. 2018;26(5):225-38

    2018

  24. [32]

    A survey of Semantic Reason- ing frameworks for robotic systems

    Liu W, Daruna A, Patel M, Ramachandruni K, Chernova S. A survey of Semantic Reason- ing frameworks for robotic systems. Robotics and Autonomous Systems. 2023;159:104294

    2023

  25. [33]

    What Computers Still Can’t Do

    Dreyfus HL. What Computers Still Can’t Do. Revised edition ed. MIT Press; 1992

    1992

  26. [34]

    Why Heideggerian Ai Failed and How Fixing It Would Require Making It More Heideggerian

    Dreyfus HL. Why Heideggerian Ai Failed and How Fixing It Would Require Making It More Heideggerian. Philosophical Psychology. 2007;20(2):247-68

    2007

  27. [35]

    Ultralytics YOLO; 2023

    Jocher G, Chaurasia A, Qiu J. Ultralytics YOLO; 2023. Available from: https://github. com/ultralytics/ultralytics

    2023

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.