RECORD
Robot Introspection via Wrench-based Action Grammars
Robotics · Submitted 16 Sep 2016
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim The contribution of our work is the presentation of a simple but generalizable semantic scheme that enables a robot to understand its high level state.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
Robotic failure is all too common in unstructured robot tasks. Despite well designed controllers, robots often fail due to unexpected events. How do robots measure unexpected events? Many do not. Most robots are driven by the senseplan- act paradigm, however more recently robots are working with a sense-plan-act-verify paradigm. In this work we present a principled methodology to bootstrap robot introspection for contact tasks. In effect, we are trying to answer the question, what did the robot do? To this end, we hypothesize that all noisy wrench data inherently contains patterns that can be effectively represented by a vocabulary. The vocabulary is generated by meaningfully segmenting the data and then encoding it. When the wrench information represents a sequence of sub-tasks, we can think of the vocabulary forming sets of words or sentences, such that each subtask is uniquely represented by a word set. Such sets can be classified using statistical or machine learning techniques. We use SVMs and Mondrian Forests to classify contacts tasks both in simulation and in real robots for one and dual arm scenarios showing the general robustness of the approach. The contribution of our work is the presentation of a simple but generalizable semantic scheme that enables a robot to understand its high level state. This verification mechanism can provide feedback for high-level planners or reasoning systems that use semantic descriptors as well. The code, data, and other supporting documentation can be found at: http://www.juanrojas.net/2017icra_wrench_introspection.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | Robotic failure is all too common in unstructured robot tasks. | background 1.00 | 0.05 | 0.02 | asserted 1.00 |
| s1 | Despite well designed controllers, robots often fail due to unexpected events. | background 0.99 | 0.07 | 0.02 | asserted 0.99 |
| s2 | How do robots measure unexpected events? Many do not. | gap 0.92 | 0.06 | 0.02 | asserted 0.96 |
| s3 | Most robots are driven by the senseplan- act paradigm, however more recently robots are working with a sense-plan-act-verify paradigm. | background 0.99 | 0.05 | 0.03 | asserted 0.99 |
| s4 | In this work we present a principled methodology to bootstrap robot introspection for contact tasks. | contribution 1.00 | 0.88 | 0.11 | asserted 0.99 |
| s5 | In effect, we are trying to answer the question, what did the robot do? | contribution 0.48 | 0.32 | 0.04 | asserted 0.54 |
| s6 | To this end, we hypothesize that all noisy wrench data inherently contains patterns that can be effectively represented by a vocabulary. | method 0.76 | 0.25 | 0.13 | speculative 0.47 |
| s7 | The vocabulary is generated by meaningfully segmenting the data and then encoding it. | method 1.00 | 0.70 | 0.05 | asserted 0.99 |
| s8 | When the wrench information represents a sequence of sub-tasks, we can think of the vocabulary forming sets of words or sentences, such that each subtask is uniquely represented by a word set. | method 0.95 | 0.29 | 0.07 | hedged possibility 0.65 |
| s9 | Such sets can be classified using statistical or machine learning techniques. | method 1.00 | 0.22 | 0.04 | asserted 0.69 |
| s10 | We use SVMs and Mondrian Forests to classify contacts tasks both in simulation and in real robots for one and dual arm scenarios showing the general robustness of the approach. | result 0.64 | 0.97 | 0.21 | asserted 1.00 |
| s11 | The contribution of our work is the presentation of a simple but generalizable semantic scheme that enables a robot to understand its high level state. | contribution 1.00 | 0.91 | 0.56 | asserted 1.00 |
| s12 | This verification mechanism can provide feedback for high-level planners or reasoning systems that use semantic descriptors as well. | interpretation 0.81 | 0.32 | 0.15 | hedged possibility 0.96 |
| s13 | The code, data, and other supporting documentation can be found at: http://www.juanrojas.net/2017icra_wrench_introspection. | contribution 0.50 | 0.29 | 0.02 | asserted 0.92 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
Robotic failure is all too common in unstructured robot tasks. Despite well designed controllers, robots often fail due to unexpected events. How do robots measure unexpected events? Many do not. Most robots are driven by the senseplan- act paradigm, however more recently robots are working with a sense-plan-act-verify paradigm. In this work we present a principled methodology to bootstrap robot introspection for contact tasks. In effect, we are trying to answer the question, what did the robot do? To this end, we hypothesize that all noisy wrench data inherently contains patterns that can be effectively represented by a vocabulary. The vocabulary is generated by meaningfully segmenting the data and then encoding it. When the wrench information represents a sequence of sub-tasks, we can think of the vocabulary forming sets of words or sentences, such that each subtask is uniquely represented by a word set. Such sets can be classified using statistical or machine learning techniques. We use SVMs and Mondrian Forests to classify contacts tasks both in simulation and in real robots for one and dual arm scenarios showing the general robustness of the approach. The contribution of our work is the presentation of a simple but generalizable semantic scheme that enables a robot to understand its high level state. This verification mechanism can provide feedback for high-level planners or reasoning systems that use semantic descriptors as well. The code, data, and other supporting documentation can be found at: http://www.juanrojas.net/2017icra_wrench_introspection.
Human review
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.