Advanced Robotics - Project
Phase 2 and 3
Harish Balasubramaniam, Sarthak Dalal, Soumya Tyagi
Bandicam Screen recording
Scaling in reward functions
Reward function | Scale |
Termination | 4 |
Distance to Target | 2 |
Orientation to Target | 0.5 |
Collision Penalty | 4 |
Time to target | 0.25 |
Distance to obstacle | -1 |
Goal achievement | 5 |