Sliding Mode UAV Control
Last edited: August 8, 2025Problem: we have a drone in an elevator; the elevator moves and hence the drone has a non-internal reference frame.
For a dynamical system which has uncertainty term which is not observable to the controller, one could control for position and velocity well with a Sliding Mode Controller
Sliding mode observer controls for the disturbance in the non-inertial reference frame.
slope (statistics)
Last edited: August 8, 2025Smith 20??
Last edited: August 8, 2025One-Liner
“impact of approximation decreases as steps from the root node”
Novelty
- combined alpha-vector and forward heuristics to guide search of belief states before backup
- 100x times faster in PBVI
- scales to huge environments
Goal: minimize “regret” (difference until optimal policy)
Novelty HSVI 2
- Projected the upper bound onto a convex hull (HSVI2: via approximate convex hull projection)
- uses blind lower bound
Notable Methods
Key Figs
New Concepts
Notes
smooth function
Last edited: August 8, 2025a function is called smoo
Social Learning
Last edited: August 8, 2025Social Learning is the property of learning with other agents in your environment.
- learning from feedback
- human ai coordination
- social learning
- adversarial training
