what have what have i learned i learned at the at the
play

WHAT HAVE WHAT HAVE I LEARNED I LEARNED AT THE AT THE DISC DISC - PowerPoint PPT Presentation

WHAT HAVE WHAT HAVE I LEARNED I LEARNED AT THE AT THE DISC DISC SUMMER SCHOOL? SUMMER SCHOOL? Christos G. Cassandras CODES Lab. - Boston University WHY HYBRID SYSTEMS ? COMPLEXITY ( PHYSICAL , BUT MOSTLY OPERATIONAL/ORGANIZATIONAL )


  1. WHAT HAVE WHAT HAVE I LEARNED I LEARNED AT THE AT THE DISC DISC SUMMER SCHOOL? SUMMER SCHOOL? Christos G. Cassandras CODES Lab. - Boston University

  2. WHY HYBRID SYSTEMS ? � COMPLEXITY ( PHYSICAL , BUT MOSTLY OPERATIONAL/ORGANIZATIONAL ) � UNCERTAINTY – Switching is one way of dealing with uncertainties, (e.g., ABS) Christos G. Cassandras CODES Lab. - Boston University

  3. (EMBEDDING, COMPLEXITY REDUCTION) ABSTRACTION HYBRID DES SYSTEMS VERIFICATION, CONTROL CONTROL HYBRID DES SYSTEMS ABSTRACTION (COMPLEXITY REDUCTION) Christos G. Cassandras CODES Lab. - Boston University

  4. QUANTIZATION HYBRID DES SYSTEMS CONTROL (STABILITY, FAULT DIAGNOSIS) Christos G. Cassandras CODES Lab. - Boston University

  5. SWITCHING VIEWPOINT � STABILITY, REACHABILITY � IDENTIFICATION SWITCH TIMING � OPTIMAL CONTROL MODE SELECTION (SCHEDULING) Christos G. Cassandras CODES Lab. - Boston University

  6. HYBRID SYSTEMS v HYBRID MODELS OF (COMPLEX) DYNAMIC SYSTEMS Quantization Gear shifting in a vehicle Christos G. Cassandras CODES Lab. - Boston University

  7. DETERMINISTIC HYBRID SYSTEMS v STOCHASTIC HYBRID SYSTEMS Not as much progress… Christos G. Cassandras CODES Lab. - Boston University

  8. EXPERIMENTAL WORK, LABORATORIES FOR HS � “REAL SYSTEMS”, LAB. PROTOTYPES � SIMULATION � LEGO SYSTEMS Christos G. Cassandras CODES Lab. - Boston University

  9. MODELS FOR HS Why is THIS model BETTER than THAT for a specific problem ? Which modeling framework should I use ? Who cares ? They are all GREAT! So, just pick one that suits you or your problem and just solve your problem… …and please try NOT to invent yet another modeling framework!... NOTE: I counted 16 new terms and 18 acronyms in one presentation… Christos G. Cassandras CODES Lab. - Boston University

  10. x x 1 1 b 1 a 1 a 1 b x < 2 y x < 2 x < 2, x ≥ 2 x ≥ 2 y y ≥ 3 2 2 2 EVENTS: x → x = 2 y → y = 3 TIMED AUTOMATON TIMED PETRI NET TIMED AUTOMATON (with clock structure) Christos G. Cassandras CODES Lab. - Boston University

  11. COMPUTER SCIENCE APPROACHES v SYSTEMS AND CONTROL APPROACHES - Build tools - Dynamic Programming - Explore/Simulate - Linear Programming - Let the computer do the work - Lagrange Relaxation [JUST DO IT] - Gradient-Based Optim. - Perturbation Analysis [JUST PROVE IT] Can we combine ? Can we combine ? Christos G. Cassandras CODES Lab. - Boston University

Recommend


More recommend