| Περίληψη: | This thesis studies all the stages involved in the perception, control and manipulation
of robotic laparoscopic tools with emphasis given to the pivot trajectories and the RCM
constrained motion planning. The forward and inverse kinematics of the 7 DoF KUKA
iiwa14 industrial robot arm was studied followed by kinematics of the attached Barrett
hand gripper for grasping purposes. Minimal Invasive Surgery (MIS) necessitates the
study of Remote Center-of-Motion (RCM) constraint for the pivot motions, the elbow-
up constraint to avoid collisions as well as the workspace constraints and singularities.
The transformation of the surgical task space into the robot’s taskspace and the joint
space is subsequently analyzed along with the robot’s manipulability. Emphasis was
given in calculating various geometric paths for the robot to follow inside the surgical
task. The equations for circular, circular arc, line segment, helical, cubic spline, b-
spline, higher-order polynomial and trapezoid and s-curve velocity profile trajectories
are studied in detail in order to generate pivot motions with a wide variety. Simulations
were conducted using the ROS framework and libraries like Gazebo, RViz and MoveIt
using the RRTConnect path planning algorithm. The simulations were evaluated with
measurements of time, position accuracy and RCM distance deviation. This thesis also
briefly studies a simple recognition of a laparoscopic tool and the estimation of its position
and orientation using computer vision as well as the calculation of three points on the
surgical tool where the gripper’s fingers will be placed in order to grasp the object with
a satisfactory force closure. Finally this thesis studies some control system schemes like
for example the RCM tracking and pivot motion control.
|
|---|
