Description

Objectives

• To generate smooth and comfortable trajectories for an autonomous vehicle
• To avoid dynamic obstacles based on sensor data
• To modify driving behaviour to follow the rules of the road

In this assignment, you will be developing a trajectory planner that can handle highway driving while surrounded by dynamic obstacles moving at various speeds. The planner will interface with a driving simulator to allow you to visualize the ego vehicle in motion.

Resources and Instructions

This assignment requires you to complete the following assignment:
https://github.com/udacity/CarND-Path-Planning-Project

Once you’ve cloned the repo, go to “src” folder, replace “main.cpp” with the one provided on
Learn, and add “spline.h” (also on Learn) to “src” folder. The two files are explained in this Q&A video by Udacity:

Instructions for how to finish the project can be found in the README. But we have modified the grading scheme so that you make get part marks even if some criteria are not met. Please see the “Marking Scheme” section and tutorial slides for new instructions.

The simulator can be downloaded here:
https://github.com/udacity/self-driving-car-sim/releases/tag/T3_v1.2

Deliverables

The deliverable for this project is a zipped project folder containing your developed code (i.e. whatever files you modified/created in “src”), as well as a writeup in txt or pdf format. Your submission should compile and run because we will be testing your code. Please do not include the simulator in your submission.

Please follow the naming convention for your zip file: a8_<user_id>.zip .

Marking Scheme

Assignments are marked on a 0-11 point scale.

Trajectory generation (4 points)
• Watch Udacity’s Q&A video, answer the following questions:
o 1) how is lane following achieved? (1 points)
o 2) how to use spline to generate a smooth trajectory? (2 point) o 3) how to avoid collision with the car in front? (0.5 points) o 4) how to avoid cold start? (0.5 points)
o Can elaborate on 2) (4-6 sentences), but for 1) 3) 4), 2 sentences maximum

Behaviour Planning (5 points)
During 2 miles of driving, achieve the following (1 point for each item):
• 1) Perform lane shift when front vehicle is too slow, done at least once.
• 2) No collisions
• 3) Count( Exceed( speed lim || acc lim || jerk lim )) <= 1 o If any limits are exceeded for more than 2 sec straight, you lose this point
• 4) Complete 2 miles within 3 min
• 5) No limits exceeded
Important note: An achievement can only be unlocked by successfully completing the previous ones, for example, if you fail 1), then you automatically lose marks for 2)-6); if you fail 2), you automatically lose marks for 3)-6) and so on.

Writeup (2 points)
• Briefly explain your approach for behaviour planning and any modifications to the provided trajectory generation code (7-10 sentences)
Policies

Collaboration
You can discuss the problem with peers, but you must design and implement your own solution independently.

Use of online resources