Trajectory Optimization based on CasADi, implemented for IIT's Centauro robot.
:heavy_check_mark: Offline/online trajectory optimization for quadruped robots and quadruped manipulators carrying heavy payload.
:heavy_check_mark: Implemented using CasADi and its interface to Ipopt solver.
:heavy_check_mark: Implemented in Python, based on ROS.
The repo has been tested on Ubuntu 18.04 and 20.04, using the XbotCore 2 and Cartesian Interface developed within the HHCM lab of IIT. To use the repo you will need:
casannis_walking is a catkin package. To install it you need to git clone
into your workspace:
git clone https://github.com/ADVRHumanoids/casannis_walking.git
Build with catkin_make
/ catkin build
The repo is not maintained to enable off-the-self deploying, but rather consists a record of the work done for the corresponding publication mentioned below. People interested are encouraged to contact the author.
For offline trajectory optimization try
mon launch casannis_walking cartesio.launch
mon launch casannis_walking gait.launch
Or
mon launch casannis_walking cartesio_with_arm_ee.launch
mon launch casannis_walking gait_payload.launch
For online trajectory optimization (under development) try
mon launch casannis_walking cartesio.launch
rosrun casannis_walking online_simple_gait_replay_node.py
mon launch casannis_walking simple_gait_online_planner.launch
The repo is related with the following publication:
Dadiotis I., Laurenzi A., Tsagarakis N., “Trajectory Optimization for Quadruped Mobile Manipulators that Carry Heavy Payload”, 2022 IEEE-RAS International Conference on Humanoid Robots (Humanoids 2022), Ginowan, Okinawa, Japan
Available: https://ieeexplore.ieee.org/document/10000252
If you want to refer to this work please cite with:
@inproceedings{dadiotis2022trajectory,
title={Trajectory Optimization for Quadruped Mobile Manipulators that Carry Heavy Payload},
author={Dadiotis, Ioannis and Laurenzi, Arturo and Tsagarakis, Nikos},
booktitle={2022 IEEE-RAS 21st International Conference on Humanoid Robots (Humanoids)},
pages={291--298},
year={2022},
organization={IEEE}
}
Ioannis Dadiotis, ioannis.dadiotis@iit.it