Adaptive Kinematic Modelling for Multiobjective Control of a Redundant Surgical Robotic Tool

Kormushev

2021

Applied Sciences

Self Cite

Robots have been predominantly controlled using conventional control methods that require prior knowledge of the robots’ kinematic and dynamic models. These controllers can be challenging to tune and cannot directly adapt to changes in kinematic structure or dynamic properties. On the other hand, model-learning controllers can overcome such challenges. Our recently proposed model-learning orientation controller has shown promising ability to simultaneously control a three-degrees-of-freedom robot manipulator’s end-effector pose. However, this controller does not perform optimally with robots of higher degrees-of-freedom nor does it resolve redundancies. The research presented in this paper extends the state-of-the-art kinematic-model-free controller to perform pose control of hyper-redundant robot manipulators and resolve redundancies by tracking and controlling multiple points along the robot’s serial chain. The results show that with more control points, the controller is able to reach desired poses in fewer steps, yielding an improvement of up to 66%, and capable of achieving complex configurations. The algorithm was validated by running the simulation 100 times, and it was found that, in 82% of the times, the robot successfully reached the desired target pose within 150 steps.

Section: Related Workmentioning

confidence: 99%

Kinematic-Model-Free Redundancy Resolution Using Multi-Point Tracking and Control for Robot Manipulation

Kormushev

2021

Applied Sciences

Self Cite

“…Different types of mechanical transmissions have been used in the design of surgical robots, with the vast majority being tendon-driven [1]. Due to the model uncertainties and the highly complex nonlinearities in tendon-driven systems, researchers employed machine learning approaches [2] such as Artificial Neural Networks [3], Gaussian Mixture Regression, K-nearest Neighbour Regression, and Extreme Machine Learning [4] to learn the system's model.…”

Section: Introductionmentioning

confidence: 99%

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

IEEE Robot. Autom. Lett.

Modugno

Lanari

et al. 2021

Self Cite

In order to guarantee precision and safety in robotic surgery, accurate models of the robot and proper control strategies are needed. Bayesian Neural Networks (BNN) are capable of learning complex models and provide information about the uncertainties of the learned system. Model Predictive Control (MPC) is a reliable control strategy to ensure optimality and satisfaction of safety constraints. In this work we propose the use of BNN to build the highly nonlinear kinematic and dynamic models of a tendon-driven surgical robot, and exploit the information about the epistemic uncertainties by means of a Hierarchical MPC (Hi-MPC) control strategy. Simulation and real world experiments show that the method is capable of ensuring accurate tip positioning, while satisfying imposed safety bounds on the kinematics and dynamics of the robot.

“…In [13], [14] feedforward Artificial Neural Networks (ANNs) were used to learn the forward kinematics of a robotic system and analytical derivation is carried out to compute the robot Jacobian for control purposes. A similar approach is used in [15], [16] where the derivation of the Jacobian is exploited to solve the redundancy problem by means of Null Space Projection [17].…”

Section: Introductionmentioning

confidence: 99%

Augmenting Loss Functions of Feedforward Neural Networks with Differential Relationships for Robot Kinematic Modelling

2021 20th International Conference on Advanced Robotics (ICAR)

Chappell

Kormushev

2021

Self Cite

Model learning is a crucial aspect of robotics as it enables the use of traditional and consolidated model-based controllers to perform desired motion tasks. However, due to the increasing complexity of robotic structures, modelling robots is becoming more and more challenging, and analytical models are very difficult to build, particularly for redundant robots. Machine learning approaches have shown great capabilities in learning complex mapping and have widely been used in robot model learning and control. Generally, inverse kinematics is learned, directly obtaining the desired control commands given a desired task. However, learning forward kinematics is simpler and allows the computation of the robot Jacobian and enables the exploitation of the optimality of controllers. Nevertheless, typical learning methods have no knowledge about the differential relationship between the position and velocity mappings. In this work, we present two novel loss functions to train feedforward Artificial Neural network (ANN) which incorporate this information in learning the forward kinematic model of robotic structures, and carry out a comparison with standard ANN training using position data only. Simulation results show that incorporating the knowledge of the velocity mapping improves the suitability of the learnt model for control tasks.