Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Autonomous Iterative Motion Learning (AI-MOLE) of a SCARA Robot for Automated Myocardial Injection

Michael Meindl, Raphael Mönkemöller, Thomas Seel

TL;DR

The results indicate that the combination of a SCARA robot and learning method achieves sufficiently precise motion to potentially enable automatic myocardial injection if similar results can be obtained in a real-world setting.

Abstract

Stem cell therapy is a promising approach to treat heart insufficiency and benefits from automated myocardial injection which requires highly precise motion of a robotic manipulator that is equipped with a syringe. This work investigates whether sufficiently precise motion can be achieved by combining a SCARA robot and learning control methods. For this purpose, the method Autonomous Iterative Motion Learning (AI-MOLE) is extended to be applicable to multi-input/multi-output systems. The proposed learning method solves reference tracking tasks in systems with unknown, nonlinear, multi-input/multi-output dynamics by iteratively updating an input trajectory in a plug-and-play fashion and without requiring manual parameter tuning. The proposed learning method is validated in a preliminary simulation study of a simplified SCARA robot that has to perform three desired motions. The results demonstrate that the proposed learning method achieves highly precise reference tracking without requiring any a priori model information or manual parameter tuning in as little as 15 trials per motion. The results further indicate that the combination of a SCARA robot and learning method achieves sufficiently precise motion to potentially enable automatic myocardial injection if similar results can be obtained in a real-world setting.

Autonomous Iterative Motion Learning (AI-MOLE) of a SCARA Robot for Automated Myocardial Injection

TL;DR

The results indicate that the combination of a SCARA robot and learning method achieves sufficiently precise motion to potentially enable automatic myocardial injection if similar results can be obtained in a real-world setting.

Abstract

Stem cell therapy is a promising approach to treat heart insufficiency and benefits from automated myocardial injection which requires highly precise motion of a robotic manipulator that is equipped with a syringe. This work investigates whether sufficiently precise motion can be achieved by combining a SCARA robot and learning control methods. For this purpose, the method Autonomous Iterative Motion Learning (AI-MOLE) is extended to be applicable to multi-input/multi-output systems. The proposed learning method solves reference tracking tasks in systems with unknown, nonlinear, multi-input/multi-output dynamics by iteratively updating an input trajectory in a plug-and-play fashion and without requiring manual parameter tuning. The proposed learning method is validated in a preliminary simulation study of a simplified SCARA robot that has to perform three desired motions. The results demonstrate that the proposed learning method achieves highly precise reference tracking without requiring any a priori model information or manual parameter tuning in as little as 15 trials per motion. The results further indicate that the combination of a SCARA robot and learning method achieves sufficiently precise motion to potentially enable automatic myocardial injection if similar results can be obtained in a real-world setting.
Paper Structure (8 sections, 19 equations, 4 figures)

This paper contains 8 sections, 19 equations, 4 figures.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. Proposed Method
  4. Model Learning
  5. Input Learning
  6. Autonomous Parameterization
  7. Simulation Results
  8. Conclusion

Figures (4)

  • Figure 1: This work considers the problem of automatically injecting stem cells into the myocardium using a robotic manipulator (A), which has to move a syringe with sub-millimeter precision. To meet this requirement, we propose Autonomous Iterative Motion Learning (AI-MOLE) for multi-input/multi-output systems and conduct a preliminary study using a SCARA robot (C), where the robot's endeffector is equipped with a syringe and the robot's two rotational degrees of freedom are utilized to position the syringe in the horizontal plane (B).
  • Figure 2: Each of AI-MOLE's learning iterations consists of three steps. (A) The current input trajectory is applied to the unknown, multi-input/multi-output dynamics yielding a state trajectory. (B) The experimental data are then used to learn a GP model of the unknown dynamics. (C) Finally, AI-MOLE utilizes the GP model in an ILC update rule to compute the input trajectory of the next trial. The learning scheme is repeated until the output trajectory converges sufficiently close to the reference trajectory.
  • Figure 3: AI-MOLE is evaluated using a simplified SCARA robot that consists of two links that rotate in the horizontal plane. AI-MOLE has to solve three different reference tracking tasks which respectively lead to dynamic motions that cover the robot's entire task space.
  • Figure 4: The simulation results of AI-MOLE learning to track three different reference trajectories. Despite the unknown, multi-input/multi-output dynamics, AI-MOLE achieves satisfying tracking performance within 10--15 trials without any manual parameter tuning, and the final motions are sufficiently precise to possibly enable automated myocardial injection.