Link: https://www.formulastudent.de/teams/fsd/
Link: https://www.formulastudent.de/fileadmin/user_upload/all/2020/rules/FS-Rules_2020_V1.0.pdf
Link: https://arxiv.org/pdf/1905.05150.pdf
Description: This paper is written by the AMZ team (won several driverless competitions). Their paper covers their entire software stack and all algorithms that they used to setup their system.
Link: https://github.com/AMZ-Driverless/fsd-resources
Description: Readme page with a collection of resources collection by AMZ.
Link: https://github.com/AMZ-Driverless/fssim
Description: Repo with their simulator. Implemented with ROS1 and gazebo.
Link: https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python
Description: Filter will be needed to locate the car within a track. This resource covers a list of filters by providing code samples in jupyter notebooks.
Link: https://github.com/matterport/Mask_RCNN
Description: One of the earliest object detection frameworks. Largely replace by object detector such as YOLO.
Link: https://github.com/ultralytics/yolov5
Description: The family of YOLO detectors were originally developed in the darknet framework. This repo however implements the latest iteration in Pytorch which is much more user friendly.
Link: https://github.com/rwightman/efficientdet-pytorch
Description: Detection framework that offers different speed/accuracy trade-offs.
Link: https://pytorch.org/
Description: Highly flexible deep learning framework. The preferred choice for this project.
Link: https://keras.io/
Description: DL framework aimed at making DL code easy.
Link: https://www.tensorflow.org/
Description: DL framework made for production use. Exists in two versions tf1 and tf2. tf1 is not user friendly while tf2 is not commonly used.
Link: https://developer.nvidia.com/drive/drive-agx
Description: Hardware that we might use.
Link: https://www.ros.org/
Description: Framework commonly used for controlling robots.