Prediction Module for data dumper compatibility check #119
Conversation
…AccelerationHeading, ConstantSpeedYawRate, ConstantManitudeAccelAndYawRate.
… best predicted trajectory among all of the proposals generated by each physics-based model
| @@ -58,6 +58,13 @@ def draw_trajetory_points(world, waypoints, z=0.25, | |||
| arrow_size=arrow_size, | |||
| color=color, | |||
| life_time=lt) | |||
| def draw_points(world, points, z, color=carla.Color(255, 102, 0), lt=0.06, size=0.05): | |||
There was a problem hiding this comment.
- [ Please write docstring for this to tell users what is the meaning of every variable and what is the data type]
| @@ -105,6 +106,9 @@ def __init__(self, vehicle, carla_map, config_yaml): | |||
| # collision checker | |||
| self._collision_check = CollisionChecker( | |||
| time_ahead=config_yaml['collision_time_ahead']) | |||
| # Warning!! | |||
| # hard coded!!! | |||
There was a problem hiding this comment.
- [ Use todo/TODO to note the things you need to change later]
| @@ -105,6 +106,9 @@ def __init__(self, vehicle, carla_map, config_yaml): | |||
| # collision checker | |||
| self._collision_check = CollisionChecker( | |||
| time_ahead=config_yaml['collision_time_ahead']) | |||
| # Warning!! | |||
| # hard coded!!! | |||
| self.prediction_manager = PredictionManager(int(4/0.05), int(3/0.05), 0.05) | |||
There was a problem hiding this comment.
You can pass the parameter fixed_delta_second from the yaml file here instead of hard coding it. Check behavior planner as an exmaple.
| def angle_diff(x, y): | ||
| """ | ||
| Get the smallest angle difference between 2 angles: the angle from y to x. | ||
| Parameters |
|
|
||
|
|
||
| class TrajectoryData: | ||
| def __init__(self, l): |
There was a problem hiding this comment.
Remember to add docstring for every class in the future.
| @@ -0,0 +1,274 @@ | |||
| # -*- coding: utf-8 -*- | |||
| """ | |||
| Physcis-based trajectory prediction model | |||
There was a problem hiding this comment.
Restructure this core script please. Rename it to physics_model, and put all functions that are not a member of any class to another script(suggested name: physics_utils). By this way, the code structure is more clear.
| return np.linalg.norm(x - y) | ||
|
|
||
| assert dist(preds, gt) < 1e-10 | ||
| return |
There was a problem hiding this comment.
no need to write return for such method.
|
Some general comment: add prediction module parameters in the yaml file so users could choose different model and set different params. |
No description provided.