Abstract
One of the most important elements of a robot’s control system is its Inverse Kinematic Model (IKM), which calculates the position and velocity references required by the robot’s actuators to follow a trajectory. The methods that are commonly used to synthesize the IKM of open-chain robotic systems strongly depend on the geometry of the analyzed robot. Those methods are not systematic procedures that could be applied equally in all possible cases. This project presents the development of a systematic procedure to synthesize the IKM of non-redundant open-chain robotic systems using Groebner Basis theory, which does not depend on the geometry of the robot’s structure. The inputs to the developed procedure are the robot’s Denavit–Hartenberg parameters, while the output is the IKM, ready to be used in the robot’s control system or in a simulation of its behavior. The Groebner Basis calculation is done in a two-step process, first computing a basis with Faugère’s F4 algorithm and a grevlex monomial order, and later changing the basis with the FGLM algorithm to the desired lexicographic order. This procedure’s performance was proved calculating the IKM of a PUMA manipulator and a walking hexapod robot. The errors in the computed references of both IKMs were absolutely negligible in their corresponding workspaces, and their computation times were comparable to those required by the kinematic models calculated by traditional methods. The developed procedure can be applied to all Cartesian robotic systems, SCARA robots, all the non-redundant robotic manipulators that satisfy the in-line wrist condition, and any non-redundant open-chain robot whose IKM should only solve the positioning problem, such as multi-legged walking robots.
Highlights
The modeling and design of a robot’s control system begins with the resolution of its kinematic problem, which is divided into two parts: the computation of the robot’s Forward Kinematics and the synthesis of the Inverse Kinematic Model (IKM).Appl
Applying the Denavit–Hartenberg’s method, with the parameters presented in Table 1, the solution to the Forward Kinematics problem of the PUMA 560 is the homogeneous transformation shown in Equation (1), which establishes the transformation between the base of the robot and the anchor point of its in-line wrist
This robot was selected as a testbench because each of its legs has three rotational degrees of freedom for positioning, like most industrial robotic arms and many multi-legged walking robots, and it only requires the resolution of the positioning kinematic problem
Summary
The modeling and design of a robot’s control system begins with the resolution of its kinematic problem, which is divided into two parts: the computation of the robot’s Forward Kinematics and the synthesis of the Inverse Kinematic Model (IKM). It is important to clarify that, while these methods solve the IKP, they do not synthesize an Inverse Kinematic Model (IKM) This implies that the position references that are offered as output do not come from a differentiable function; these procedures are not capable of computing the speed or acceleration references for the robot’s actuators. This work objective is to use Groebner Basis theory to develop a systematic procedure for the synthesis of the IKM of non-redundant open-chain robotic systems.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
