Launching the Planner¶
In order to use the planner, integrate the graph_ltpl package in your code. In order to get started, we provide two example scripts. These scripts demonstrate the code integration and allow to test the planners features.
Minimal Example¶
A minimal example of the local trajectory planner is parametrized in the ‘main_min_example.py’ script in the root directory. Within this example, the planner is executed without any interfaces (e.g. object list) and no configured logging. By default a live-visualization is shown (note: the live-visualization slows down the execution drastically). Launch the code with the following command:
python3 main_min_example.py
Note
When a certain track configuration is executed the first time, a offline graph is generated first. This will take some time (progress bar is displayed). On completion, a plot of the generated graph with all its edges is displayed (in case the visualization is enabled). After closing the figure with the offline graph, a live visualization is launched and the vehicle starts driving.
Standard Example¶
A more comprehensive example of the local trajectory planner is given in the ‘main_std_example.py’ script (also in the root directory). Within this example, the planner is executed with basic interfaces:
- Dummy object list - another vehicle driving with reduced speed along the racing-line (NOTE: here executed without any prediction of the object-vehicle’s motion)
- Dummy blocked zone - a small region of the track is blocked for the ego vehicle (useful to block certain regions, e.g. pit lane or dirty track)
- Logging to file - the environment and planned trajectories of every time-stamp are logged to a file, which can be visualized with the interactive log-viewer afterwards
- Prioritized action selection - instead of a behavior planner, in this example any ‘pass’ action is executed, when available
By default a live-visualization is shown (note: the live-visualization slows down the execution drastically). Launch the code with the following command:
python3 main_std_example.py
Note
When a certain track configuration is executed the first time, a offline graph is generated first. This will take some time (progress bar is displayed). On completion, a plot of the generated graph with all its edges is displayed (in case the visualization is enabled). After closing the figure with the offline graph, a live visualization is launched and the vehicle starts driving.
Further Steps¶
A description for usage of the planner class in your project is given in Using the Planner Package. Furthermore, the parameterization, development tools and the log visualizer is tackled in that Chapter.