FusionAD
FusionAD copied to clipboard
SJSU-AD
An open source autonomous driving stack by San Jose State University Autonomous Driving Team
## Problem Autoware's [calibration tool](https://github.com/CPFL/Autoware/wiki/Calibration) expects the Lidar to publish to "/points_raw", but our current driver publishes to "/velodyne_point_cloud" ## Solution Remap "/velodyne_point_cloud" to "/points_raw" in [the decoder launch file](https://github.com/SJSU-AD/FusionAD/blob/develop/src/utility/velodyne_puck/velodyne_puck_decoder/launch/velodyne_puck_decoder_nodelet.launch)...
- [ ] Note (in README in Arduino and Localization?) that baud rate needs to be changed in multiple locations. State exact file names and location of change within each...
- [Validator](https://github.com/SJSU-AD/FusionAD/blob/test_deployment/src/modules/geodesy/src/validate_enu_to_latlon.py) - [CSV Converter](https://github.com/SJSU-AD/FusionAD/blob/test_deployment/src/modules/geodesy/src/convert_enu_to_latlon.py)
- Define different levels of lidar-based obstacle detection - Establish modes of detection - Develop handshake between perception and manager module
The manager module will define vehicle action mode based on the current state of the vehicle as well as its certainty and perception information.
When the points within the virtual box meets the threshold and there is a human detected, play a configurable MP3 tone.
1. Set up sensors, vehicle, and world TF. 2. Set up odom information 3. Figure out a mapping solution
There should be a seamless way to disengage the autonomy mode and recover. @joeyzhu00 and @nathansrinivasan5 please work together with me on brainstorming a solution for this issue.
The controller produces a lot of small adjustment in the steering wheel even when the vehicle is driving in a straight line. Investigate the root cause and provide a solution...