REal-time Whole-body Motion Planning for Mobile MANIpulators Using Environment-adaptive Search and Spatial-temporal Optimization
- Jan 29, 2024: REMANI-Planner is accepted to ICRA 2024.
REMANI-Planner presents a motion planning method capable of generating high-quality, safe, agile and feasible trajectories for mobile manipulators in real time.
Authors: Chengkai Wu*, Ruilin Wang*, Mianzhi Song, Fei Gao, Jie Mei and Boyu Zhou
Institutions: SYSU STAR Group, HITSZ MAS Lab and ZJU FAST Lab.
Paper: Real-time Whole-body Motion Planning for Mobile Manipulators Using Environment-adaptive Search and Spatial-temporal Optimization, 2024 IEEE International Conference on Robotics and Automation (ICRA).
@INPROCEEDINGS{10610192,
author={Wu, Chengkai and Wang, Ruilin and Song, Mianzhi and Gao, Fei and Mei, Jie and Zhou, Boyu},
booktitle={2024 IEEE International Conference on Robotics and Automation (ICRA)},
title={Real-time Whole-body Motion Planning for Mobile Manipulators Using Environment-adaptive Search and Spatial-temporal Optimization},
year={2024},
volume={},
number={},
pages={1369-1375},
keywords={Service robots;Dynamics;Transportation;Real-time systems;Planning;Safety;Complexity theory},
doi={10.1109/ICRA57147.2024.10610192}}
If you find this work useful or interesting, please kindly give us a star ⭐, thanks!😀
Compiling tests passed on Ubuntu 20.04 with ROS installed.
- ROS (tested with Noetic)
sudo apt install libompl-dev libeigen3-dev
cd /usr/include
sudo ln -sf eigen3/Eigen Eigen
sudo ln -sf eigen3/unsupported unsupported
cd ${your catkin workspace}/src
git clone -b master --single-branch https://github.com/SYSU-STAR/REMANI-Planner.git
cd ..
catkin_make -DCMAKE_BUILD_TYPE=Release
- Navigating in dense cuboids map
source devel/setup.bash
roslaunch remani_planner exp0.launch
You should see the simulation in rviz. You can use the 2D Nav Goal
to send a trigger to start navigation.
- Navigating through a bridge
source devel/setup.bash
roslaunch remani_planner exp1.launch
- Make the following adjustments in the
remani_planner/mm_config/src/mm_config.cpp
file:- Modify the
getAJointTran
function to calculate the homogeneous transformation between different frames of the manipulator. - Modify the
setLinkPoint
function to set the position of collision spheres in the respective frame. - Modify the
getMMMarkerArray
function based on the urdf file to generate the marker array for MM visualization.
- Modify the
- Adapt the
mm_param.yaml
file located in theremani_planner/remani_planner/config/
directory to configure parameters specific to your MM.
Note: We have provided an example for the UR5 in our code. To use the UR5, simply follow these steps:
- Open the
mm_param.yaml
file located in theremani_planner/plan_manage/config/
directory. - Locate the
parameter
forFastArmer
and comment it out by adding a "#" symbol at the beginning of the line. - Uncomment the
parameter
forUR5
by removing the "#" symbol at the beginning of the line. - You are now ready to conduct above experiments with a mobile base incorporating the UR5 configuration.
We use MINCO as our trajectory representation.
We borrow the framework from AutoTrans.
We would like to thank colleagues at Huawei for their support for this work: Zehui Meng and Changjin Wang.
The source code is released under the GPLv3 license.
For any technical issues, please contact Chengkai Wu([email protected]) or Ruilin Wang([email protected]).