Gerard Canal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Gerard Canal has a total of 46 co-authors. Here are the top co-authors by number of shared publications:
- Guillem Alenyà
Institut de Robòtica i Informàtica Industrial, CSIC-UPC
8 shared publications - Lennart Wachowiak
King’s College London, London, United Kingdom
7 shared publications - Oya Celiktutan
King’s College London, London, United Kingdom
6 shared publications - Andrew Coles
Informatics, King’s College London, London, United Kingdom of Great Britain and Northern Ireland
5 shared publications - Carme Torras
Institut de Robòtica i Informàtica Industrial, CSIC-UPC
5 shared publications Carme Torras
CSIC-UPC, Institut de Robòtica i Informàtica Industrial, C/ Llorens i Artigas 4-6, Barcelona, 08028, Spina
2 shared publications
Latest Publications
- Open Access
- 12 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
Goal Generation for Autonomous Planning Agents
Generating objectives and reasoning about them is key for intelligent planning agents to become more autonomous. Once a set of achievable goals has been chosen, a symbolic planner can find an executable plan to fulfil them. However, finding these goals is a complex problem when the aim is to optimise conflicting metrics such as goal benefit, plan cost, and its makespan. In this paper, we first propose a method to identify a set of potential goals in a temporal PDDL domain. Next, we find the achievable subset of goals that yields the most benefit by posing the problem as an integer linear program. Finally, a planner is used to solve the resulting subset of goals. The proposed method is generic and planner-independent, generating sets of goals that are neither provably unreachable nor mutually exclusive, thereby increasing the likelihood of finding a solvable plan. We demonstrate the effectiveness of the proposed approach in comparison with a baseline oversubscription planner.
- 12 November 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Intelligent Systems and Technology
Open AccessGoal Generation for Autonomous Planning Agents
Generating objectives and reasoning about them is key for intelligent planning agents to become more autonomous. Once a set of achievable goals has been chosen, a symbolic planner can find an executable plan to fulfil them. However, finding these goals is a complex problem when the aim is to optimise conflicting metrics such as goal benefit, plan cost, and its makespan. In this paper, we first propose a method to identify a set of potential goals in a temporal PDDL domain. Next, we find the achievable subset of goals that yields the most benefit by posing the problem as an integer linear program. Finally, a planner is used to solve the resulting subset of goals. The proposed method is generic and planner-independent, generating sets of goals that are neither provably unreachable nor mutually exclusive, thereby increasing the likelihood of finding a solvable plan. We demonstrate the effectiveness of the proposed approach in comparison with a baseline oversubscription planner.
- 23 October 2026
- Published by Springer Nature
- p. 643-662
TP-ONT: An Ontology for Enhanced Reasoning in AI Task Planning via Atomic Decomposition
- 23 October 2026
- Published by Springer Nature
- p. 643-662
TP-ONT: An Ontology for Enhanced Reasoning in AI Task Planning via Atomic Decomposition
- Open Access
- 10 October 2026
- Published by Springer Nature
- p. 423-441
User Preferences for Enhanced Robot Adaptation in Assistive Tasks
- 10 October 2026
- Published by Springer Nature
- p. 423-441
Open AccessUser Preferences for Enhanced Robot Adaptation in Assistive Tasks
- Open Access
- 25 September 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Human-Robot Interaction
- Vol. 15 (5) , 1-25
What Questions Should Robots Be Able to Answer? A Dataset of User Questions for Explainable Robotics
With the growing use of large language models and conversational interfaces in human–robot interaction, robots’ ability to answer user questions is more important than ever. We therefore introduce a dataset of 1,893 user questions for household robots, collected from 100 participants and organized into 12 categories and 70 subcategories. Most work in explainable robotics focuses on why-questions, such as “Why did you clean the bathroom but not the kitchen?” In contrast, our dataset provides a wide variety of questions, from questions about simple execution details to questions about how the robot would act in hypothetical scenarios—thus giving roboticists valuable insights into what questions their robot needs to be able to answer. To collect the dataset, we created 15 video stimuli and 7 text stimuli, depicting robots performing varied household tasks. We then asked participants on Prolific what questions they would want to ask the robot in each portrayed situation. In the final dataset, the most frequent categories are questions about task execution details (21.4%), the robot’s capabilities (12.6%), and performance assessments (10.7%). Although questions about how robots would handle potentially difficult scenarios and ensure correct behavior are less frequent, users rank them as the most important for robots to be able to answer. Moreover, we find that users who identify as novices in robotics ask different questions than more experienced users. Novices are more likely to inquire about simple facts, such as what the robot did or the current state of the environment. As robots enter environments shared with humans and language becomes central to giving instructions and interaction, this dataset provides a valuable foundation for (i) identifying the information robots need to log and expose to conversational interfaces, (ii) benchmarking question–answering modules, and (iii) designing explanation strategies that align with user expectations.
- 25 September 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Human-Robot Interaction
- Vol. 15 (5) , 1-25
Open AccessWhat Questions Should Robots Be Able to Answer? A Dataset of User Questions for Explainable Robotics
With the growing use of large language models and conversational interfaces in human–robot interaction, robots’ ability to answer user questions is more important than ever. We therefore introduce a dataset of 1,893 user questions for household robots, collected from 100 participants and organized into 12 categories and 70 subcategories. Most work in explainable robotics focuses on why-questions, such as “Why did you clean the bathroom but not the kitchen?” In contrast, our dataset provides a wide variety of questions, from questions about simple execution details to questions about how the robot would act in hypothetical scenarios—thus giving roboticists valuable insights into what questions their robot needs to be able to answer. To collect the dataset, we created 15 video stimuli and 7 text stimuli, depicting robots performing varied household tasks. We then asked participants on Prolific what questions they would want to ask the robot in each portrayed situation. In the final dataset, the most frequent categories are questions about task execution details (21.4%), the robot’s capabilities (12.6%), and performance assessments (10.7%). Although questions about how robots would handle potentially difficult scenarios and ensure correct behavior are less frequent, users rank them as the most important for robots to be able to answer. Moreover, we find that users who identify as novices in robotics ask different questions than more experienced users. Novices are more likely to inquire about simple facts, such as what the robot did or the current state of the environment. As robots enter environments shared with humans and language becomes central to giving instructions and interaction, this dataset provides a valuable foundation for (i) identifying the information robots need to log and expose to conversational interfaces, (ii) benchmarking question–answering modules, and (iii) designing explanation strategies that align with user expectations.
- Open Access
- 16 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 222-227
Neurosymbolic Explanation Selection in Robotics: Combining the Strengths of Planning and Foundation Models for XAI
- 16 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 222-227
Open AccessNeurosymbolic Explanation Selection in Robotics: Combining the Strengths of Planning and Foundation Models for XAI
- Open Access
- 16 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 1372-1374
Workshop on Designing Transparent and Understandable Robots (D-TUR)
In order for robots to act as an empowering tool in today's society, it is necessary for users to understand what those robots can and cannot do. This can happen through the design of the robot, through external information such as signs and manuals, or through the interaction itself. With this workshop, we aim to provide a platform for participants who approach robot transparency from different angles to share and discuss their experiences. Participants will have the opportunity to deepen their knowledge of transparent and understandable robot interactions through two keynotes and paper presentations. The diverse backgrounds of participants will be brought together in interactive exercises that aim to address questions about topics such as how transparency can be used to reduce inequalities and empower society and other open problems in the field. The guided break-out sessions aim at giving participants the chance to reflect on the keynotes, paper presentations and their own work and to share those reflections with participants with other perspectives.
- 16 March 2026
- Published by Association for Computing Machinery (ACM)
- p. 1372-1374
Open AccessWorkshop on Designing Transparent and Understandable Robots (D-TUR)
In order for robots to act as an empowering tool in today's society, it is necessary for users to understand what those robots can and cannot do. This can happen through the design of the robot, through external information such as signs and manuals, or through the interaction itself. With this workshop, we aim to provide a platform for participants who approach robot transparency from different angles to share and discuss their experiences. Participants will have the opportunity to deepen their knowledge of transparent and understandable robot interactions through two keynotes and paper presentations. The diverse backgrounds of participants will be brought together in interactive exercises that aim to address questions about topics such as how transparency can be used to reduce inequalities and empower society and other open problems in the field. The guided break-out sessions aim at giving participants the chance to reflect on the keynotes, paper presentations and their own work and to share those reflections with participants with other perspectives.
- Open Access
- 23 February 2026
- Published by Elsevier in Information Sciences
- Vol. 741
Ontological foundations for contrastive explanatory narration of robot plans
- 23 February 2026
- Published by Elsevier in Information Sciences
- Vol. 741
Open AccessOntological foundations for contrastive explanatory narration of robot plans
- Open Access
- 11 November 2025
- Published by Springer Nature in International Journal of Social Robotics
- Vol. 17 (11) , 2853-2854
Special Issue on Prediction and Anticipation for Reasoning in Human-Robot Interaction
- 11 November 2025
- Published by Springer Nature in International Journal of Social Robotics
- Vol. 17 (11) , 2853-2854
Open AccessSpecial Issue on Prediction and Anticipation for Reasoning in Human-Robot Interaction
- Open Access
How Far Are LLMs from Symbolic Planners? An NLP-Based Perspective
The reasoning and planning abilities of Large Language Models (LLMs) have been a frequent topic of discussion in recent years. Their ability to take unstructured planning problems as input has made LLMs' integration into AI planning an area of interest. Nevertheless, LLMs are still not reliable as planners, with the generated plans often containing mistaken or hallucinated actions. Existing benchmarking and evaluation methods investigate planning with LLMs, focusing primarily on success rate as a quality indicator in various planning tasks, such as validating plans or planning in relaxed conditions. In this paper, we approach planning with LLMs as a natural language processing (NLP) task, given that LLMs are NLP models themselves. We propose a recovery pipeline consisting of an NLP-based evaluation of the generated plans, along with three stages to recover the plans through NLP manipulation of the LLM-generated plans, and eventually complete the plan using a symbolic planner. This pipeline provides a holistic analysis of LLM capabilities in the context of AI task planning, enabling a broader understanding of the quality of invalid plans. Our findings reveal no clear evidence of underlying reasoning during plan generation, and that a pipeline comprising an NLP-based analysis of the plans, followed by a recovery mechanism, still falls short of the quality and reliability of classical planners. On average, only the first 2.65 actions of the plan are executable, with the average length of symbolically generated plans being 8.4 actions. The pipeline still improves action quality and increases the overall success rate from 21.9% to 27.5%.
Open AccessHow Far Are LLMs from Symbolic Planners? An NLP-Based Perspective
The reasoning and planning abilities of Large Language Models (LLMs) have been a frequent topic of discussion in recent years. Their ability to take unstructured planning problems as input has made LLMs' integration into AI planning an area of interest. Nevertheless, LLMs are still not reliable as planners, with the generated plans often containing mistaken or hallucinated actions. Existing benchmarking and evaluation methods investigate planning with LLMs, focusing primarily on success rate as a quality indicator in various planning tasks, such as validating plans or planning in relaxed conditions. In this paper, we approach planning with LLMs as a natural language processing (NLP) task, given that LLMs are NLP models themselves. We propose a recovery pipeline consisting of an NLP-based evaluation of the generated plans, along with three stages to recover the plans through NLP manipulation of the LLM-generated plans, and eventually complete the plan using a symbolic planner. This pipeline provides a holistic analysis of LLM capabilities in the context of AI task planning, enabling a broader understanding of the quality of invalid plans. Our findings reveal no clear evidence of underlying reasoning during plan generation, and that a pipeline comprising an NLP-based analysis of the plans, followed by a recovery mechanism, still falls short of the quality and reliability of classical planners. On average, only the first 2.65 actions of the plan are executable, with the average length of symbolically generated plans being 8.4 actions. The pipeline still improves action quality and increases the overall success rate from 21.9% to 27.5%.
- Open Access
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 657-665
A Time Series Classification Pipeline for Detecting Interaction Ruptures in HRI Based on User Reactions
To be able to react to interaction ruptures such as errors, a robot needs a way of realizing such a rupture occurred. We test whether it is possible to detect interaction ruptures from the user’s anonymized speech, posture, and facial features. We showcase how to approach this task, presenting a time series classification pipeline that works well with various machine learning models. A sliding window is applied to the data and the continuously updated predictions make it suitable for detecting ruptures in real-time. Our best model, an ensemble of MiniRocket classifiers, is the winning approach to the ICMI ERR@HRI challenge. A feature importance analysis shows that the model heavily relies on speaker diarization data that indicates who spoke when. Posture data, on the other hand, impedes performance. Our code is available online1.
- 4 November 2024
- Published by Association for Computing Machinery (ACM)
- p. 657-665
Open AccessA Time Series Classification Pipeline for Detecting Interaction Ruptures in HRI Based on User Reactions
To be able to react to interaction ruptures such as errors, a robot needs a way of realizing such a rupture occurred. We test whether it is possible to detect interaction ruptures from the user’s anonymized speech, posture, and facial features. We showcase how to approach this task, presenting a time series classification pipeline that works well with various machine learning models. A sliding window is applied to the data and the continuously updated predictions make it suitable for detecting ruptures in real-time. Our best model, an ensemble of MiniRocket classifiers, is the winning approach to the ICMI ERR@HRI challenge. A feature importance analysis shows that the model heavily relies on speaker diarization data that indicates who spoke when. Posture data, on the other hand, impedes performance. Our code is available online1.