Sign in Subscribe
Machine learning

Robotics Researchers Develop Integrated Framework for Manipulating Deformable Linear Objects

Strategic Insights

— 1 min read

Researchers at the University of Edinburgh have developed an integrated framework for the Real2Sim2Real problem of manipulating deformable linear objects (DLOs) based on visual perception. The framework uses likelihood-free inference (LFI) to compute the posterior distributions for the physical parameters of DLOs, allowing for the simulation of their behavior. The researchers demonstrated the utility of this approach by deploying sim-trained DLO manipulation policies in the real world in a zero-shot manner. They also studied the implications of the resulting domain distributions in sim-based policy learning and real-world performance.

Key Takeaways:

  • The research team developed an integrated framework for manipulating DLOs based on visual perception, which uses LFI to compute posterior distributions for the physical parameters of DLOs.
  • The framework was tested by deploying sim-trained DLO manipulation policies in the real world in a zero-shot manner, without any further fine-tuning.
  • The researchers evaluated the capacity of a prominent LFI method to perform fine classification over the parametric set of DLOs, using only visual and proprioceptive data obtained in a dynamic manipulation trajectory.
  • The study concluded that the integrated framework has implications for sim-based policy learning and real-world performance.
  • Funders for this research include the EPSRC, as part of the CDT in RAS through Heriot-Watt University, University of Edinburgh, and the UKRI Turing AI World Leading Researcher Fellowship on AI for Person-Centred and Teachable Autonomy.
  • The research was peer-reviewed and published in the IEEE Robotics and Automation Letters.
  • The study was conducted by researchers at the University of Edinburgh's School of Informatics.

Statistics:

  • The research was published in the IEEE Robotics and Automation Letters, Volume 10, Issue 8, pp. 8075-8082.
  • The study evaluated the capacity of a popular LFI method to perform fine classification over the parametric set of DLOs.
  • The framework was tested using a set of DLOs with different physical parameters.

Sources:

  • Distributional Treatment of Real2sim2real for Object-centric Agent Adaptation In Vision-driven Dlo Manipulation. IEEE Robotics and Automation Letters, 2025;10(8):8075-8082.
  • IEEE Robotics and Automation Letters, Ieee-inst Electrical Electronics Engineers Inc, 445 Hoes Lane, Piscataway, NJ 08855-4141, USA.
  • NewsRx. Data on Robotics and Automation Detailed by Researchers at University of Edinburgh (Distributional Treatment of Real2sim2real for Object-centric Agent Adaptation In Vision-driven Dlo Manipulation). Robotics & Machine Learning. August 4, 2025; p 87.

Read more