AutonomousRobot
This project simulates our path-planning for an autonomous robot that tries to find a way to reach a goal (target) in certainly environment
Usage:
python Robot_run.py -n <number of run times> -m <map name> -w <worldname> -r vision_range -sx <x> -sy <y> -gx <x> -gy <y>
- n: number of run times
- < 0 or 0: run until meet the given goal
- n: number of run times
- default: 1
- m: input map name, default _map.csv
- w: input world model, no default.
- r: robot's vision range.
- sx, sy: start point of (x,y), type = float, default = 0.0, 0.0
- gx, gy: goal point of (x,y), type = float, default = 50.0, 50.0
- p: picking strategy
- g: picking global first
- l for picking local first,
- default: l (picking local first)
- rank_type: ranking type: ranking based on RRTtreeStar for or distance_and_angle formula
- r: ranking by RRTreeStar
- da: ranking by distance_and_angle
- default: r
Examples:
Ranking by distance and angle for formula, picking_global_set_first strategy:
python Robot_run.py -n 5 -p g -rank_type da -m _MuchMoreFun.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0 -p
python Robot_run.py -n 0 -p g -rank_type da -m _map.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python Robot_run.py -n 0 -p g -rank_type da -w _world.png -sx 5 -sy 10 -gx 250 -gy 310 -r 40
Ranking by distance and angle for formula, picking_local_set_first strategy
python Robot_run.py -n 5 -p l -rank_type da -m _MuchMoreFun.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python Robot_run.py -n 0 -p l -rank_type da -m _map.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python Robot_run.py -n 0 -p l -rank_type da -w _world.png -sx 5 -sy 10 -gx 250 -gy 310 -r 40
Ranking by RRTreeStar, picking_local_set_first strategy
python Robot_run.py -n 5 -p l -rank_type r -m _MuchMoreFun.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python Robot_run.py -n 0 -p l -rank_type r -m _map.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python Robot_run.py -n 0 -p l -rank_type r -w _world.png -sx 5 -sy 10 -gx 250 -gy 310 -r 40
To run demo for RRTx algorithm:
python RRTree_X.py -m _map_blocks_1.csv
python RRTree_X.py -m _map_blocks_2.csv
python RRTree_X.py -n 5 -m _MuchMoreFun.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python RRTree_X.py -n 0 -m _map.csv -sx 5 -sy 5 -gx 35.0 -gy 50.0
python RRTree_X.py -n 0 -w _world.png -sx 5 -sy 10 -gx 250 -gy 310 -r 40
To see animation of RRT/RRTstart algorithms (without obstacle(s) ):
- ss: sample size (default 2000)
python RRTree.py -ss 500
python RRTree_star.py -ss 500
For obstacle(s), i will update later if i have time
To run demo for the assumption of An and Hoai's Theory:
python Robot_theory.py -n 0 -r 100 -m _forest.csv -gx 500 -gy 500
python Robot_theory.py -n 0 -r 90 -m _forest.csv -gx 500 -gy 500
python Robot_theory.py -n 0 -r 80 -m _forest.csv -gx 500 -gy 500
- Set robot_vision parameter (option -r ) to see the different outcomes of experiments
To run experiment
python Easy_experiment.py -n 100 -m _forest.csv
python Easy_experiment.py -n 10 -w _world.png
recommend setting the following parameters in the program_config file to skip animation/printing.
easy_experiment = False
save_image = True
The results of experiment are:
- result<date_time>.csv file: contains all infomation in text form.
- pdf, images file: are plots at final step.
to visualize results of experiments:
python Easy_experiment_lib.py -r result<date_time>.csv
To generate a map:
Usage:
python map_generator.py -n <number of obstacles> -m <map name> -img <from_world_image>
- mn: input map name, default _map_temp.csv
- n: number of obstacles.
Example 1 (generating map from user input):
python map_generator.py -n 5 -m _map_temp.csv
- Click on the given plot to input points
- Middle mouse click to turn next obstacle. Each obstacle contains a maximum of 100000 vertices
Example 2 (generating map from image):
python map_generator.py -img _world.png
From world image 
to map data (csv) 
To display a map:
python map_display.py -m <map name>
Example: python map_display.py -m _map_temp.csv