Completed
How does the polynomial optimization work?
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
UW CSE Robotics - Nicholas Roy, "Planning to Fly and Drive Aggressively"
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Challenges
- 3 Fast Approximation Techniques For Planning Aggressive Flight
- 4 Planning with Complex Dynamics
- 5 The problem with the RRT
- 6 Differential flatness
- 7 Differentially flat representations exist for many systems
- 8 RRT* as a initialization
- 9 Performance Comparison
- 10 How does the polynomial optimization work?
- 11 Unknown Environment Assumptions
- 12 Different Cost Functions
- 13 Collecting Training Data
- 14 Learning Collision Probabilities
- 15 Planning With Collision Probabilities
- 16 What is the biggest thing holding back the performance?
- 17 Generalization
- 18 Modelling distributions
- 19 Enforcing smoothness
- 20 Theorem
- 21 Generalized Kernel Estimation
- 22 Our solution: A Bayesian approach
- 23 Empirical vs Guaranteed Safety
- 24 Safety guarantee emphasizes sensing
- 25 Receding-horizon planning
- 26 Strategy: Predict a correction to the shortest-path heuristic
- 27 Training: Brief sketch
- 28 Data Representation Descriptive Features
- 29 Results: Characteristics of Learned Model
- 30 Autonomous RC Car Experiments
- 31 Experiment: Baseline Planner
- 32 Experiment: Our Planner
- 33 What's Next...?
- 34 What is Needed?
- 35 Acknowledgements
