Author: cursious92

Pre-operative Offline Optimization of Insertion Point Location for Safe and Accurate Surgical Task Execution

Link

Accepted for publication in IEEE IROS 2021-International Conference on Intelligent Systems and Robots

Description:

The optimization framework

In this work we present an optimization approach to identify the optiaml insertion point location for laparascopic surgeies when employing macro-micro robotic manipulators.

Our results show how the location of the Remote Center of Motion (RCM) afects the robot’s perfomance.

The cost function takes into account safety in avoiding nearby organs, maximizing robot’s dexterity, and ensuring precision in perfoming a specified task (path tracking for tumor resection).

Our cost function and resilence to error strategy ensure the optimal location resides in a region with small variations in the perfomance in case of deviations from the optimum.

Results showing the obstacle avoidance and the effects of different RCM locations

Bayesian Neural Network Modeling and Hierarchical MPC for a Tendon-Driven Surgical Robot with Uncertainty Minimization

Link

Accepted for publication in Robotics and Automation Letters (RAL) and ICRA 2021- International Conference on Robotics and Automation.

Description:

The Hi-MPC controller

In this work we present a Hierarchical Model Predictive Control (Hi-MPC) approach in conjunction with Bayesian Neural Networks (BNN) to model and control a tendon-driven surgical robotic tool.

BNN are here employed because they allow obtaining not only the model prediciton, but also the associated model uncertainty.

The Hi-MPC is instead used to solve two prioritized subtasks:

  • primary task: tracking a desired path;
  • secondary task: ezploit the robot redundnacy and minimize the model’s uncertainty so as to reconfigure itself in more reliable state space regions.

Our simulation and real world experiments show that the uncertainty minimization helps imrpoving the control performance and make the path tracking more accurate.

Comparison of the control with only the primary MPC and the Hi-MPC approach.

Model predictive control for a tendon-driven surgical robot with safety constraints in kinematics and dynamics

Link

Accepted for publication in IROS 2020- International Conference on Intelligent Robots and Systems.

Description:

The MPC controller

In this work we evaluate the applicatbility of Model Predictive Control (MPC) to control a tendon-driven surgical robotic tool. The dynamic model of the robot is learnt by means of Artificial Neural Networks (ANN) and used to impose bounds on the dynamics of the system.

We evaluate our approach on different bounding values.

Tracking results in simulation tests

A Novel Approach for Outlier Detection and Robust Sensory Data Model Learning

Link

Accepted for publication in IROS 2019- International Conference on Intelligent Robots and Systems

Description:

The robust algorithm for nonlinear regression with ANN

In this work we developed a novel algorithm for nonlinear regression with Artificial Neural Networks (ANN) capable of rejecting outliers in the dataset and thus improve the modelling.