Dynamical Programming
References
Motivation
- Solve the multi-stage optimization problem
DP in finite-horizon discrete-time problem
Principle of Optimality
DP algorithm
- Backward and forward computation
- Proof of the algorithm
Remarks - Value function
Example: double integrator
Results
Video demo of the motion of the double integrator, with grid dimension 201*201
Example Code
Matlab | |
---|---|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
|
Observations
- Increase the grid size will get more closed results
- No matter how the grid size is changed, the optimal ctrl always chooses from -1,0,1, like a bang-bang control
Example: double integrator in minimum-time problem
DP in infinite-horizon discrete-time problem
Bellman equation
DP in continuous-time problem
placeholder