evenator / carnd-capstone Goto Github PK

we downloaded a lot of udacity and LISA data, we need to convert it into a different format to train the tensorflow models. If we get it working, we will then post the format we want the data in from the simulator

In the tl_detector we need to load two different models based on the situation. Can someone figure out if there is a variable to read, or how to tell which model we should load?

Now that these files are stable, we can add them to the repository so that we can submit the project.

Test the complete self-driving car stack prior to submission.

Once your waypoint updater is publishing /final_waypoints, the waypoint_follower node will start publishing messages to the /twist_cmd topic. At this point, you have everything needed to build the dbw_node. After completing this step, the car should drive in the simulator, ignoring the traffic lights.

Detect the traffic light and its color from the /image_color. Udacity has provided a bag file to test with.

Once you have correctly identified the traffic light and determined its position, you can convert it to a waypoint index and publish it.

Detect the traffic light and its color from the /image_color. The topic /vehicle/traffic_lights contains the exact location and status of all traffic lights in simulator, so you can test your output.

Use /traffic_waypoint to change the waypoint target velocities before publishing to /final_waypoints. Your car should now stop at red traffic lights and move when they are green.

Complete a partial waypoint updater which subscribes to /base_waypoints and /current_pose and publishes to /final_waypoints.

Does this work?
Stop and restart PID controllers depending on the state of /vehicle/dbw_enabled

Reported by @alanswx on Slack:

FYI - right now we have a crash bug when we play back the rosbag - i forget whether it is in the waypoint updater or dbw