1 / 7

Spring 2004 DGC Planning Team Meeting - Week 3

Spring 2004 DGC Planning Team Meeting - Week 3. Planning team: Jason Yosinski jason@its Will Coulter wkc5@its Ryan Cable ryanc@its Will Heltsley will@caltech Jason Raycroft raycroft@its Sue Ann Hong sueh@its Haomiao Huang haomiao@its Jack Lee jackl@its Advisor: Lars Cremean lars@cds

peak
Télécharger la présentation

Spring 2004 DGC Planning Team Meeting - Week 3

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Spring 2004 DGC Planning Team Meeting - Week 3 • Planning team: • Jason Yosinski jason@its • Will Coulter wkc5@its • Ryan Cable ryanc@its • Will Heltsley will@caltech • Jason Raycroft raycroft@its • Sue Ann Hong sueh@its • Haomiao Huang haomiao@its • Jack Lee jackl@its • Advisor: • Lars Cremean lars@cds • Consultants: • Les White leswhite@adelphia.net • Tony Fender afender@its Wed, April 14, 2004

  2. Meeting Goals and Agenda • Goals: • Review and update GOTChA chart • Brief status update from team members • Outline together several system architectures, focused on flexibility • Draft schedule for rest of term, including any data collection/analysis, and documentation • Agenda: • 4:00 - Meeting goals and agenda (assign note taker) • 4:05 - GOTChA chart update • 4:15 - Status updates • 4:25 - System architectures discussion • 4:45 - Action items, term schedule • 5:00 - Adjourn Wed, April 14, 2004

  3. Goals (DGC II) 40 mph waypoint following in (TBD) environment Objectives (by end of term) Research, outline, and document several path planning architectures Define reusable code structure for future development Pursue development and documentation of existing road-following algorithms Collect information and sensory data from course route Technical Challenges Current path (arc) evaluation code, coupled with reaction delay, is too slow Required obstacle detection range depends on reaction time and speed Approach Division of efforts: Optimal path generation (H, WH) Path planning architectures (SAH) Road following (KD, JG) Static map game plan (RC, AF, LW) DEM  Cost map  D* (JY, WC) Principal component analysis (H, JR) Roughness/clutter estimation, for mode management (JR) Spring 2004 DGC Planning team GOTChA chart Wed, April 14, 2004

  4. Spring 2004 DGC Planning team status chart • Status (division of efforts): • Optimal path generation (H, WH): Looking into code • Path planning architectures (SAH): Lit. search • Road following (KD, JG): Getting up to speed, reading (papers and code) • Static map game plan (RC, AF, LW): Investigating ArcView • DEM  Cost map  D* (JY, WC): Nothing yet • Principal component analysis (H, JR): Talking with Lyle • Roughness/clutter estimation, for mode management (JR): Nothing yet One pair Firely cameras calibrated Image capture code written, not integrated One pair Firely cameras calibrated Image capture code written, not integrated Owner HELP! Needs work Interface Working Wed, April 14, 2004

  5. 1 2 3 Wk 1 Wk 2 Wk 3 Wk 4 Wk 5 Wk 6 Wk 7 Wk 8 Wk 9 Wk 10 LBC: 1/19 tests - final code structure sufficiently far apart and numerous waypoints Planning Team Timeline 4 3/29 4/5 4/12 4/19 4/26 5/3 5/10 5/17 5/24 5/31 Team  Sandie Fender ES demo Project meetings: Mon, 7-8 pm  Event Team meetings: Wed., 4pm-5pm                      ? ? Wed, April 14, 2004

  6. Arch. 1: Arbiter Framework Pros: Versatile, easy to integrate mode management Cons: Reactive, not deliberative Arch. 2: Optimal planning over DEMs Pros: Deliberative approach, can plan dynamically feasibile (aggressive) trajectories with NTG Cons: Computationally expensive Q: How to fit single image data into this approach? Dependencies: Sensor capabilities, in terms of detection capabilities, range, etc. Planning Architectures Wed, April 14, 2004

  7. Individual actions: Create phpBB discussion thread for system architectures (LC) Reading (papers, code), searches on individual parts (Everyone) Next week agenda items: Decisions to make: system architecture, new sensors investigation? Status update Decide on interim documentation structure Division of efforts: Optimal path generation (H, WH) Path planning architectures (SAH) Road following (KD, JG) Static map game plan (RC, AF, LW) DEM  Cost map  D* (JY, WC) Principal component analysis (H, JR) Roughness/clutter estimation, for mode management (JR) Action Items Wed, April 14, 2004

More Related