Search-Based Task Planning with Learned Skill Effect Models for Lifelong Robotic Manipulation

Liang, Jacky; Sharma, Mohit; Alex, LaGrassa,; Vats, Shivam; Saxena, Saumya; Kroemer, Oliver

doi:10.1109/icra46639.2022.9811575

“…We also assume the domain is given in the form of the fact types and lifted operators, with their preconditions, effects, and controllers. Techniques from machine learning, combined with this planning framework, could be used to relax these assumptions and enable wider generalizations (Konidaris et al, 2018; Wang et al, 2021; Silver et al, 2021; Liang et al, 2022).…”

Section: Discussionmentioning

confidence: 99%

Robust planning for multi-stage forceful manipulation

Holladay,

Lozano-Pérez,

Rodriguez

2023

The International Journal of Robotics Research

2

0

View full text Add to dashboard Cite

Multi-step forceful manipulation tasks, such as opening a push-and-twist childproof bottle, require a robot to make various planning choices that are substantially impacted by the requirement to exert force during the task. The robot must reason over discrete and continuous choices relating to the sequence of actions, such as whether to pick up an object, and the parameters of each of those actions, such how to grasp the object. To enable planning and executing forceful manipulation, we augment an existing task and motion planner with constraints that explicitly consider torque and frictional limits, captured through the proposed forceful kinematic chain constraint. In three domains, opening a childproof bottle, twisting a nut and cutting a vegetable, we demonstrate how the system selects from among a combinatorial set of strategies. We also show how cost-sensitive planning can be used to find strategies and parameters that are robust to uncertainty in the physical parameters.

show abstract

“…Nonetheless, similar recent frameworks or pipelines to the one this paper presents do exist. For example, Sharma et al [11] use reinforcement learning to develop and compose taskaxis robot controllers in place of skills, and Liang et al [12] open up the skill sequence for the learning agent to solve force-controlled manipulation tasks. While the structure and sequence of skills remain fixed in our work, the framework we present incorporates a formal layer of skill composition that facilitates straightforward interchangeability and potentially allows non-expert human access, both of which are desirable for an industrial application.…”

Section: Considered Problem and Related Researchmentioning

confidence: 99%

Simulation-based Learning of the Peg-in-Hole Process Using Robot-Skills

Lämmle

¹

,

Tenbrock

²

,

Bálint

³

et al. 2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Increasingly volatile markets challenge companies and demand flexible production systems that can be quickly adapted to new conditions. Machine Learning has proven to show significant potential in supporting the human operator during the time-consuming and complex task of robot programming by identifying relevant parameters of the underlying robot control program. We present a solution to learn these parameters for contact-rich, force-controlled assembly tasks from a simulation using hardware-independent robot skills. We show that successful learning and real-world execution are possible even under process deviation and tolerances utilizing the designed learning system. We present learning skill parameters as high-level robot control, evaluation and comparison of extensive simulations, and preliminary experiments on a physical robot test-bed. The developed solution approach is evaluated and discussed using the Peg-in-Hole process, a typical benchmark process in force-controlled assembly.

show abstract

“…We also assume the domain is given in the form of the fact types and lifted operators, with their preconditions, effects and controllers. Techniques from machine learning, combined with this planning framework, could be used to relax these assumptions and enable wider generalizations (Konidaris et al 2018;Wang et al 2021;Silver et al 2021;Liang et al 2022).…”

Section: Future Directionsmentioning

confidence: 99%

Planning for Multi-stage Forceful Manipulation

Holladay

¹

,

Lozano-Pérez

²

,

Rodríguez

³

2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Multi-step forceful manipulation tasks, such as opening a push-and-twist childproof bottle, require a robot to make various planning choices that are substantially impacted by the requirement to exert force during the task. The robot must reason over discrete and continuous choices relating to the sequence of actions, such as whether to pick up an object, and the parameters of each of those actions, such how to grasp the object. To enable planning and executing forceful manipulation, we augment an existing task and motion planner with constraints that explicitly consider torque and frictional limits, captured through the proposed forceful kinematic chain constraint. In three domains, opening a childproof bottle, twisting a nut and cutting a vegetable, we demonstrate how the system selects from among a combinatorial set of strategies. We also show how cost-sensitive planning can be used to find strategies and parameters that are robust to uncertainty in the physical parameters.

show abstract

Search-Based Task Planning with Learned Skill Effect Models for Lifelong Robotic Manipulation

Cited by 14 publications

References 24 publications

Robust planning for multi-stage forceful manipulation

Robust planning for multi-stage forceful manipulation

Simulation-based Learning of the Peg-in-Hole Process Using Robot-Skills

Planning for Multi-stage Forceful Manipulation

Contact Info

Product

Resources

About