Feedback Control Theory: Architectures and Tools for Real-Time Decision Making II
summary
Control design emphasizes concrete choices: PID controllers dominate (claimed PID ≈95% use, PI ≈94%), with the integral term removing constant disturbances; state‑space design uses MATLAB’s place command or LQR to pick K and auto‑tuning to set gains; trajectory generation uses receding‑horizon / MPC computed every 10–100 Hz over 10s horizons (demo runs at 100 Hz on past $30k hardware, now phone‑feasible). Reactive synthesis for temporal specs uses GR(1) with BDDs (Omega/DD) yielding cubic algorithms on large state spaces.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 0:43Control Systems
- 1:08Design Patterns
- 1:35PID Control
- 4:59Integral Feedback
- 9:41Robustness
- 11:01Feedback Design
- 14:29Reachability Test
- 15:40Control Design
- 17:39Choice of Feedback
- 21:07Constraint Optimization
- 24:55The Inner Loop
- 30:06Re receding rise and control
- 32:58Design process
- 35:59Design iterations
- 39:30Feedback loops
- 43:57Temporal logic
- 48:17System specifications
- 57:34Summary
related talks
57mReal-Time Decision Making in Hybrid Systems
15mOptimal Control for Electromagnetic Haptic Guidance Systems
49mUsing Deep Learning for Perception in Autonomous Systems: A Perspective from Control Theory
19mAirways: Optimization-Based Planning of Quadrotor Trajectories according to High-Level User Goals
59mOn Human-robot Interactions via Multimodal Intent Modeling and Reachability-based Safety Assurance
28mPreference Learning from Minimal Human Feedback for Interactive
30mOn the Aspects of AI and Robotic Autonomy
57mFrom Shop Floor Control to Physical AI: Bridging Industrial Connectivity and Intelligent Automation