A learning algorithm for visual pose estimation of continuum robots

Abstract

Abstract-Continuum robots offer signicant advantages for surgical intervention due to their down-scalability, dexterity, and structural flexibility. While structural compliance offers a passive way to guard against trauma, it necessitates robust methods for online estimation of the robot configuration in order to enable precise position and manipulation control. In this paper, we address the pose estimation problem by applying a novel mapping of the robot configuration to a feature descriptor space using stereo vision. We generate a mapping of known features through a supervised learning algorithm that relates the feature descriptor to known ground truth. Features are represented in a reduced sub-space, which we call eigenfeatures. The descriptor provides some robustness to occlusions, which are inherent to surgical environments, and the methodology that we describe can be applied to multi-segment continuum robots for closed-loop control. Experimental validation on a single-segment continuum robot demonstrates the robustness and efficacy of the algorithm for configuration estimation. Results show that the errors are in the range of 1 • .

Learning autonomous behaviours for the body of a flexible surgical robot

“…Moreover, the robot is primarily meant to be an instrument to be employed by a surgeon, who has a great expertize in practicing very complicated operations that can hardly be replaced by a well trained or programmed robot. For this reason, machine learning is usually applied to some repeated and standard parts of the surgical procedure such as suturing or retracting organs (Van Den Berg et al, 2010) or to surgical image recognition and configuration identification (Reiter et al, 2011). The current paper, instead, uses learning as a tool to make the robot capable of better assisting the surgeon during the operation and allowing him/her to perform operations that would be very difficult by using other tools.…”

Learning-based configuration estimation of a multi-segment continuum robot

“…By discretely sampling a continuous space of configurations, we can interpolate a feature descriptor manifold, which is parameterized by the configuration angles e k . In [20] we described the following forward mapping:…”

“…In this paper we extend a previous algorithm [20] to estimate the configuration of a single-segment continuum robot. This method interpolates a smooth functional mapping from robot configurations to features using a look-up table.…”

Path planning based on Q-learning and three-segment method for aircraft fuel tank inspection robot

“…Several researchers have proposed methods about continuum robot path planning. The traditional methods for manipulator path planning include rapid expansion of random tree [2][3], polynomial interpolation [4], artificial potential field method [5], probability roadmap method [6], machine learning [7][8], etc. In reference [9], the authors evaluated four trajectory generation strategies in terms of the resulting Cartesian paths and spatial extent of the course of motion.…”