CS4278/CS5478 Lab 1 Motion Planning for Mobile Robots

Simulator Installation and Setup

In Lab 1pre, you have already installed ROS Kinetic. The FoodTutrle simulator for this lab is adapted from the ROS TurtleBot Stage package. Install TurtleBot Stage:

sudo apt install ros-kinetic-turtlebot-stage
sudo apt install ros-kinetic-joint-state-publisher-gui

Clone this repo and set up the environment:

git clone https://github.com/AdaCompNUS/CS4278-5478-MotionPlanning
cd CS4278-5478-MotionPlanning
catkin_make
source devel/setup.bash

If you use the ROS docker, you may need to make a docker commit.

Check your setup:

roslaunch planner turtlebot_in_stage.launch

You should see RViz and ROS Stage.

Code Overview

The FoodTurtle system consists of three components: simulator, planner, and controller. The code in this repo provides most of these components, including the required functions to set up ROS nodes, data communication, and robot control in the simulator. You will implement the planning algorithms and collision avoidance capability by filling in the template marked with # TODO: FILL ME! in base_planner.py. We recommend that you use base_planner.py as a base class and then impelement your planners as derived classes.

Code Execution

Launch the simulator and execute the planner:

roscd planner/src
sh run.sh [your map name] start_x start_y start_theta
python your_planner.py --goal 'goal_x,goal_y' --com use_com_1_map_or_not

For example, load map1.png and set the robot start pose as (x=1, y=1, θ=0):

roscd planner/src
sh run.sh map1.png 1 1 0

Set the robot goal as (x'=5, y'=5) and run the planner:

python your_planner.py --goal '5,5' --com 0

The flag --com indicates whether the COM1 map is used, as it requires a special set of environment parameters.

Debugging and Visualization

For visualization, we recommend ROS stage. RViz provides 2.5-D visualization, but may be noisy due to ROS asynchronous communication delays.

Performance Evaluation

A dataset of 5 maps are provided for evaluation, including 4 handcrafted maps (map1.png to map4.png) and a simplified COM1 level-1 floorplan (com1.jpg). Each map has a corresponding list of testcases, with the goals specified here. The robot always starts with pose (1, 1, 0).

To reduce your work, only map4.png is possily misaligned. You may assume that all other maps are sufficiently accurate.

Evaluate your planners on all testcases under the three models, DSDA, CSDA, and DSPA. For DSDA and CSDA, save the control actions in a txt files and name it {task}{map}{goal}.txt. For DSPA, save the MDP control policy in a json files and name it {task}{map}{goal_x}_{goal_y}.json. For example,

• DSDA_map2_5_5.txt for DSDA on map2.png with [5, 5] as the goal.
• CSDA_map3_9_9.txt for CSDA on map3.png with [9, 9] as the goal.
• DSPA_com1_43_10.json for the MDP policy on com1.jpg with [43, 10] as the goal.

You may find the functions to save the results in base_planner.py. Some example control files can be found here.

Submission

Submit a single zip archive MatricNo-Lab1.zip, which contains 3 directories:

• Code: your code
• Controls: the control files for all testcases
• Report: your report