Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
119 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

SkiROS2: A skill-based Robot Control Platform for ROS (2306.17030v1)

Published 29 Jun 2023 in cs.RO and cs.AI

Abstract: The need for autonomous robot systems in both the service and the industrial domain is larger than ever. In the latter, the transition to small batches or even "batch size 1" in production created a need for robot control system architectures that can provide the required flexibility. Such architectures must not only have a sufficient knowledge integration framework. It must also support autonomous mission execution and allow for interchangeability and interoperability between different tasks and robot systems. We introduce SkiROS2, a skill-based robot control platform on top of ROS. SkiROS2 proposes a layered, hybrid control structure for automated task planning, and reactive execution, supported by a knowledge base for reasoning about the world state and entities. The scheduling formulation builds on the extended behavior tree model that merges task-level planning and execution. This allows for a high degree of modularity and a fast reaction to changes in the environment. The skill formulation based on pre-, hold- and post-conditions allows to organize robot programs and to compose diverse skills reaching from perception to low-level control and the incorporation of external tools. We relate SkiROS2 to the field and outline three example use cases that cover task planning, reasoning, multisensory input, integration in a manufacturing execution system and reinforcement learning.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (31)
  1. R. Volpe, I. Nesnas, T. Estlin, D. Mutz, R. Petras, and H. Das, “The claraty architecture for robotic autonomy,” in 2001 IEEE Aerospace Conference Proceedings (Cat. No. 01TH8542), vol. 1.   IEEE, 2001, pp. 1–121.
  2. S. Bensalem, M. Gallien, F. Ingrand, I. Kahloul, and N. Thanh-Hung, “Designing autonomous robots,” IEEE Robotics & Automation Magazine, vol. 16, no. 1, pp. 67–77, 2009.
  3. M. Stenmark and J. Malec, “Knowledge-Based Industrial Robotics,” Frontiers in Artificial Intelligence and Applications, pp. 265–274, 2013.
  4. ——, “Knowledge-based instruction of manipulation tasks for industrial robotics,” Robotics and Computer-Integrated Manufacturing, vol. 33, pp. 56–67, Jun. 2015.
  5. M. Tenorth and M. Beetz, “KNOWROB — knowledge processing for autonomous personal robots,” in 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems, Oct. 2009, pp. 4261–4266.
  6. ——, “KnowRob: A knowledge processing infrastructure for cognition-enabled robots,” The International Journal of Robotics Research, vol. 32, no. 5, pp. 566–590, Apr. 2013.
  7. F. Rovida, M. Crosby, D. Holz, A. S. Polydoros, B. Großmann, R. P. Petrick, and V. Krüger, “Skiros—a skill-based robot control platform on top of ros,” Robot Operating System (ROS) The Complete Reference (Volume 2), pp. 121–160, 2017.
  8. D. Kortenkamp, R. Simmons, and D. Brugali, “Robotic systems architectures and programming,” Springer handbook of robotics, pp. 283–306, 2016.
  9. M. Mayr, F. Ahmad, K. Chatzilygeroudis, L. Nardi, and V. Krueger, “Combining planning, reasoning and reinforcement learning to solve industrial robot tasks,” IROS 2022 Workshop on Workshop on Trends and Advances in Machine Learning and Automated Reasoning for Intelligent Robots and Systems, 2022.
  10. ——, “Skill-based multi-objective reinforcement learning of industrial robot tasks with planning and knowledge integration,” in 2022 IEEE International Conference on Robotics and Biomimetics (ROBIO).   IEEE, 2022, pp. 1995–2002.
  11. F. Ahmad, M. Mayr, E. A. Topp, J. Malec, and V. Krueger, “Generalizing behavior trees and motion-generator (btmg) policy representation for robotic tasks over scenario parameters,” 2022 IJCAI Planning and Reinforcement Learning Workshop, 2022.
  12. A. Olivares-Alarcos, D. Beßler, A. Khamis, P. Goncalves, M. K. Habib, J. Bermejo-Alonso, M. Barreto, M. Diab, J. Rosell, J. Quintas et al., “A review and comparison of ontology-based approaches to robot autonomy,” The Knowledge Engineering Review, vol. 34, p. e29, 2019.
  13. C. Paxton, A. Hundt, F. Jonathan, K. Guerin, and G. D. Hager, “Costar: Instructing collaborative robots with behavior trees and vision,” in 2017 IEEE international conference on robotics and automation (ICRA).   IEEE, 2017, pp. 564–571.
  14. J. O. Huckaby, “Knowledge transfer in robot manipulation tasks,” Ph.D. dissertation, Georgia Institute of Technology, 2014.
  15. S. Balakirsky, Z. Kootbally, T. Kramer, A. Pietromartire, C. Schlenoff, and S. Gupta, “Knowledge driven robotics for kitting applications,” Robotics and Autonomous Systems, vol. 61, no. 11, pp. 1205–1214, 2013.
  16. J. L. Wyatt, A. Aydemir, M. Brenner, M. Hanheide, N. Hawes, P. Jensfelt, M. Kristan, G.-J. M. Kruijff, P. Lison, A. Pronobis et al., “Self-understanding and self-extension: A systems and representational approach,” IEEE Transactions on Autonomous Mental Development, vol. 2, no. 4, pp. 282–303, 2010.
  17. C. Paxton, Y. Barnoy, K. Katyal, R. Arora, and G. D. Hager, “Visual robot task planning,” in 2019 international conference on robotics and automation (ICRA).   IEEE, 2019, pp. 8832–8838.
  18. P. Schillinger, S. Kohlbrecher, and O. Von Stryk, “Human-robot collaborative high-level control with application to rescue robotics,” in 2016 IEEE International Conference on Robotics and Automation (ICRA).   IEEE, 2016, pp. 2796–2802.
  19. D. Stampfer, A. Lotz, M. Lutz, and C. Schlegel, “The smartmdsd toolchain: An integrated mdsd workflow and integrated development environment (ide) for robotics software,” Journal of Software Engineering for Robotics (JOSER), vol. 7, no. 1, pp. 3–19, 2016.
  20. M. Iovino, E. Scukins, J. Styrud, P. Ögren, and C. Smith, “A survey of behavior trees in robotics and ai,” Robotics and Autonomous Systems, vol. 154, p. 104096, 2022.
  21. F. Rovida, B. Grossmann, and V. Krüger, “Extended behavior trees for quick definition of flexible robotic tasks,” in 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).   IEEE, 2017, pp. 6793–6800.
  22. V. Krueger, A. Chazoule, M. Crosby, A. Lasnier, M. R. Pedersen, F. Rovida, L. Nalpantidis, R. Petrick, C. Toscano, and G. Veiga, “A vertical and cyber–physical integration of cognitive robots in manufacturing,” Proceedings of the IEEE, vol. 104, no. 5, pp. 1114–1127, 2016.
  23. H. Bruyninckx, “Open robot control software: the orocos project,” in 2001 IEEE International Conference on Robotics and Automation (ICRA), vol. 3.   IEEE, 2001, pp. 2523–2528.
  24. M. Quigley, B. Gerkey, K. Conley, J. Faust, T. Foote, J. Leibs, E. Berger, R. Wheeler, and A. Ng, “Ros: an open-source robot operating system,” in 2009 IEEE International Conference on Robotics and Automation (ICRA) Workshop on Open Source Robotics, Kobe, Japan, May 2009.
  25. S. Bøgh, O. S. Nielsen, M. R. Pedersen, V. Krüger, and O. Madsen, “Does your robot have skills?” in Proceedings of the 43rd international symposium on robotics.   VDE Verlag GMBH, 2012.
  26. F. Rovida, D. Wuthier, B. Grossmann, M. Fumagalli, and V. Krüger, “Motion generators combined with behavior trees: A novel approach to skill modelling,” in 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).   IEEE, 2018, pp. 5964–5971.
  27. C. Aeronautiques, A. Howe, C. Knoblock, I. D. McDermott, A. Ram, M. Veloso, D. Weld, D. W. SRI, A. Barrett, D. Christianson et al., “Pddl the planning domain definition language,” Technical Report, Tech. Rep., 1998.
  28. P. Eyerich, R. Mattmüller, and G. Röger, “Using the context-enhanced additive heuristic for temporal and numeric planning,” in Nineteenth International Conference on Automated Planning and Scheduling, 2009.
  29. B. Grossmann, F. Rovida, and V. Kruger, “Continuous close-range 3d object pose estimation,” in 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).   IEEE, 2019, pp. 2861–2867.
  30. D. Wuthier, F. Rovida, M. Fumagalli, and V. Krüger, “Productive multitasking for industrial robots,” in 2021 IEEE International Conference on Robotics and Automation (ICRA).   IEEE, 2021, pp. 12 654–12 661.
  31. M. Mayr, C. Hvarfner, K. Chatzilygeroudis, L. Nardi, and V. Krueger, “Learning skill-based industrial robot tasks with user priors,” in 2022 IEEE 18th International Conference on Automation Science and Engineering (CASE).   IEEE, 2022, pp. 1485–1492.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (3)
  1. Matthias Mayr (22 papers)
  2. Francesco Rovida (2 papers)
  3. Volker Krueger (17 papers)
Citations (12)
View on Semantic Scholar

Summary

Overview of SkiROS2: A Skill-Based Robot Control Platform for ROS

The paper "SkiROS2: A skill-based Robot Control Platform for ROS" presents a comprehensive framework designed to address the challenges in modern autonomous robotic systems, particularly in industrial scenarios where flexibility and interoperability are critical. The system, SkiROS2, aims to enhance robot control through a skill-based approach that integrates knowledge representation, task planning, and execution within the Robot Operating System (ROS) environment. This paper is pertinent for researchers interested in developing advanced robotic control architectures that are scalable and adaptable to diverse industry needs.

Core Contributions

SkiROS2 introduces a novel control architecture that leverages a layered, hybrid approach, amalgamating task planning and reactive execution. This structure is realized through the extended Behavior Tree (eBT) model, which facilitates effective task-level planning and execution in dynamic environments. The paper outlines key components of SkiROS2, including:

  1. Skill Model and Management: SkiROS2 employs a skill modularity approach where both primitive and compound skills can be defined. Skills are parametric procedures characterized by pre-, hold-, and post-conditions that manage their execution, allowing for parameter inference and dynamic adaptation based on the world model (WM).
  2. World Model Integration: Knowledge is structured within an RDF graph, supporting reasoning capabilities across various tasks and entities. This integration promotes a clear, explicit formulation of knowledge, enabling efficient knowledge sharing and representation.
  3. Task Manager and Vertical Integration: The task manager automates the generation of planning domains in PDDL without manual intervention, streamlining the process of task planning. This integration is crucial for interoperability with higher-level systems such as MES, enhancing the robot's autonomy and decision-making capabilities.
  4. Multi-Robot Orchestration: The architecture supports the orchestration of multiple robotic entities, maintaining a coherent world state and thus enabling complex tasks in collaborative environments.

Implications and Challenges

SkiROS2 exemplifies a robust framework for handling the demands of Industry 4.0, addressing both the need for interoperability and breaking down barriers associated with implicit knowledge representation. By emphasizing explicit knowledge structures, SkiROS2 facilitates greater flexibility in robot control and task adaptation.

From a practical standpoint, this platform allows for the seamless integration of various robotic hardware and software, leveraging ROS's extensive libraries and community support. The user interface simplifies interactions for end-users, while developers benefit from its modular and extensible design.

Numerical Results and Use Cases

The paper presents three diverse industrial scenarios to demonstrate SkiROS2's efficacy. The scenarios highlight key features such as task-level planning with visual feedback integration, dual-arm coordination for precise manipulation tasks, and the incorporation of reinforcement learning for task adaptability. These use cases illustrate how SkiROS2's modularity and planning capabilities can be employed in real-world applications, showcasing its potential for scalability and efficiency.

Future Directions

Looking ahead, the development of SkiROS2 within the ROS 2 environment is anticipated to enhance real-time performance and robustness. Additionally, the ongoing incorporation of machine learning techniques presents opportunities to further extend the platform's capabilities, particularly in learning from experiences and adapting to new tasks autonomously.

Conclusion

The SkiROS2 platform offers a sophisticated, skill-based approach to robot control, aligning with current industrial automation trends and promoting advancements in autonomous robotic systems. Its unique combination of layered hybrid control, explicit knowledge representation, and ROS integration positions it as a valuable asset for researchers and practitioners seeking to implement advanced robotic solutions in industrial settings.

Overall, SkiROS2 serves as a comprehensive framework for developing intelligent robotic systems capable of performing complex tasks with high dexterity and flexibility, and it provides a strong foundation for further enhancements in robot control architectures.

File Document Download Save Streamline Icon: https://streamlinehq.com PDF File Document Download Save Streamline Icon: https://streamlinehq.com Markdown
Github Logo Streamline Icon: https://streamlinehq.com

GitHub

  1. GitHub - RVMI/skiros2: A skill-based platform for ROS v.2 (204 stars)
Youtube Logo Streamline Icon: https://streamlinehq.com

YouTube