KinIn is a framework for iterative kinematic analysis of mechanical systems. It evaluates the feasibility of robotic calculations for complex mechanisms, particularly those with more than five degrees of freedom in heavy machinery. The framework uses iterative computation methods rather than traditional analytical techniques to achieve results.
This repository contains the framework files in Structured Text. The MATLAB directory includes a complementary example for better visualization and exploration of the mechanism. This repository is organized into four main directories: Main_project, Matlab, RaspberryPi, and Tests.
The Tests directory contains test suites and examples for all functions, including comparisons with common manual inverse kinematics calculations. The Main_project directory contains only the primary project. The PLC_PRG executes inverse and forward kinematics calculations for positions A and B as described in the paper. The RaspberryPi directory houses a specialized Raspberry Pi project that closely resembles the main project.
The functions are categorized as follows:
- Inverse Functions: Inv, LuDecomposition, LuInverse, pInv
- Linear Algebra: Eye, isValueNegligible, MatMul, Mat_byScalar, norm, Trace
- Jacobian functions: AdjointTransform, angVelToSkewSymMatrix, BodyJacobian, expToRotAxisAndAngle, Exp_se3_fromTrans, Exp_so3_fromRot, fastInverseTransform, getFwdKinematics_Body, getInvKinematics_Body, RotSO3_fromExpso3, se3ToSpatialVel, so3ToAngularVelocity, spatialVelToSE3Matrix, TransSE3_fromExp_se3, TransSE3_ToRotAndPos
For complete documentation and usage examples, please visit the KinIn_documentation.
The recommended environment is CODESYS V3.5 SP20. Installation of the OSCAT BASIC 3.3.4 library is required for the 'KinIn_Test' project. The OSCAT BASIC library package is included in this directory. For MATLAB, the code requires the Robotics System Toolbox.
- Install CODESYS V3.5 from the official website if not already installed
- Open the 'KinIn.project' in CODESYS KinIn can be integrated directly through the Library Manager. Installation requires placing the '.library' file in the Library Repository under the Tools tab.
- Install MATLAB or use MATLAB Online
- Install the Robotics System Toolbox via the addons tab
- The MATLAB folder contains all necessary files for running the demonstration
- Install the Raspberry Package to CODESYS
- Download additional libraries if it is required in the Library Manager.
The framework is designed to integrate complex workflows. All files and functions are explicitly provided in this repository. Parameters such as joint count or maximum size for inverse matrix calculations can be modified in the 'Library Parameters' tab within the library manager.
The example in the CODESYS project works as follows: The PLC_PRG runs all the examples. First it calls the example programs for linear algebra, inverse and jacobian functions. Then it runs the example positions A and B shown in the paper.
Run 'plot_robot.m' to visualize the results and verify that they match the CODESYS output, providing insight into how the mechanism will behave in real-world applications.
The framework includes examples for all functions and a real-world mechanism test case.
The MATLAB folder contains:
- STL meshes in the 'meshes' folder
- URDF file for robot description
- 'import_robot.m' file (the positions can be easily modified by changing the angles)
- Screenshots of example positions