- doc
- model
- report: Texfile of final report
- schematics
- src
- cvxgen: Codes for cvxgen
- data: experiment data record
- examples: Source codes for project.
- saved_runs: previous version of source codes
- simulation: matlab simulation for MPC
- matlab's
.m
file: Data process to generate figures
The CVXGEN is used to solve the optimization problem of MPC. If a new CVXGEN code is generated, you should use make
to generate new .o files in the cvxgen folder. Then, copy the following folder to the source file of libfranka:
- util.o, util. c
- solver.c, solver.h, solver.o
- ldl.c, ldl.o
- matrix_support.c, matrix_support.o
Notice: Do rewrite the solver.h for your own program! If you are programming with C++, the function in solver.c should be extended to C++ with extern "C"
.
Libfranka is already installed on the PC of robot lab connected to Panda.
All the source codes in src/examples
should be copied to libfranka/examples
before running the program. The final program is 0_MPC_velocity_openloop.cpp
. If any new programs are added to be compiled, you need to modify the CMakeLists.txt
.
To compile the codes, run make
under the folder libfranka\build
or libfranka\build\examples
.
You can run the program under the directory above in the terminal typing ./FileName
. The IP of the robot is 172.16.0.2
, which can be neglected when using our program. But it is required to run the default programs of Panda.