Conversation
…ison query 9cc7db18) Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/14fe05f5-d334-4c79-bb80-127b60ace689 Co-authored-by: sgbaird <45469701+sgbaird@users.noreply.github.com>
…table delimiters Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/14fe05f5-d334-4c79-bb80-127b60ace689 Co-authored-by: sgbaird <45469701+sgbaird@users.noreply.github.com>
|
@copilot for any edison queries that were vague, use your API key to get the actual artifacts (JSON I think) that have more details about the publications and authors I think. Fetch any unfetched queries from this pull request. The BO folks from "Authors appearing ≥2× across" are largely out of scope. It needs to be grounded in people who actually publish in the journals mentioned and similar journals. |
…thors (drop BoTorch/Ax tooling pool) Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/7b232848-20ca-4856-9059-a642f8db0689 Co-authored-by: sgbaird-alt <71569943+sgbaird-alt@users.noreply.github.com>
Done in 798693c.
|
|
@copilot considering #47 (especially the edison queries), who would be the best to reach out to about "lander"-style designs and studies? |
…jectories Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/deb9992b-47de-4c3b-8019-63ef76c412e2 Co-authored-by: sgbaird-alt <71569943+sgbaird-alt@users.noreply.github.com>
Added as new §3d′ "Lander"-style designs / drag-free impact survival in Top 3 to actually reach out to first:
Next tier (4–10): Vytas SunSpiral (NIAC SUPERball PI, formerly NASA Ames; now Stoke Space / RGo Robotics), Andrew Zhang + Brian Cera (UC Berkeley, Agogino lineage — best instrumented public dataset, 10–20 m drops, 235 g peak), Massimo Vespignani (SUPERball v2 design, 2018), Skelton / Sultan (TAMU / VT — class-1 lander theory), Jamshid Bayandor (Virginia Tech CRASH Lab — scaled-up TANDEM lander 2017), Jing Zhang et al. (HIT — Aerospace 2025 class-II spherical tensegrity), Madhumati Anand (biodegradable tensegrity, 75 m drops onto pavement, 2022). Note: this cluster has essentially zero overlap with the crutch-tip / assistive-device pool — only Rimoli + Pajunen are shared with the JMD/SMS pools — so if the egg-drop demo is folded into the manuscript, expect the AE to recruit a space-systems / aerospace-mechanisms reviewer from §3d′ in addition to the §3b panel. |
Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/deb9992b-47de-4c3b-8019-63ef76c412e2 Co-authored-by: sgbaird-alt <71569943+sgbaird-alt@users.noreply.github.com>
|
@copilot send an edison query - what aspects would be best to seek feedback/help about? Consider all PRs in this repo. For example, tech transfer, what some of the real-world challenges are, any immediate "gotchas" that come to mind. Draft just a couple sentences each. I'll probably throw those sentences away and rewrite, but might as well. Is there also some way they could contribute collaboratively that comes to mind? For example, external validation. |
…treach-topics query (task f18aca01) Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/3d6aa12e-ad0b-4b49-9092-fc63f9ef2682 Co-authored-by: sgbaird-alt <71569943+sgbaird-alt@users.noreply.github.com>
…ers + per-contact routing + stretch ideas) Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/3d6aa12e-ad0b-4b49-9092-fc63f9ef2682 Co-authored-by: sgbaird-alt <71569943+sgbaird-alt@users.noreply.github.com>
…es asks to 12 named contacts) Agent-Logs-Url: https://github.com/vertical-cloud-lab/tensegrity-optimization/sessions/3d6aa12e-ad0b-4b49-9092-fc63f9ef2682 Co-authored-by: sgbaird-alt <71569943+sgbaird-alt@users.noreply.github.com>
Done in 56f58de. Submitted + fetched Edison Also drafted throwaway 1–2-sentence pitches in new §3d″ "First-contact talking points" in
Edison also returned an 8-item stretch section of mechanisms I hadn't drafted: digital-twin round-robin (NIST AM-Bench style); materials-genome integration of PETG–TPU bond data; citizen-science egg-drop campaign with downloadable STL/3MF kit; prestress-monitoring IoT (smart tensegrity); topology-optimization collab with Panozzo's PolyFEM team; ASTM F42/D20 standards pathway for multi-material FFF interface bond; FTO patent landscape scan for the joint geometries; high-speed-video + accelerometer sharing consortium. Per-contact ask routing in Edison's §(e) table covers all 12 named lander-pool contacts, each anchored to one of their own peer-reviewed papers (e.g. Rimoli → Garanger et al. 2021 Soft tensegrity systems for planetary landing; Mueller → Zha et al. 2024 IEEE/ASME T-Mech; Skelton → Williamson & Skelton 2003 General class of tensegrity structures; Jing Zhang → Zhang et al. 2025 Aerospace; Anand → Anand et al. 2022 arXiv 2212.11625) — use those citations directly in the email body. |
There was a problem hiding this comment.
Pull request overview
Adds a reviewer/AE targeting memo and outreach talking points for the JMD manuscript thread (PR #20) and the egg-drop/lander outreach thread (PR #47), grounded in committed Edison LITERATURE_HIGH artifacts and a reproducible Edison submission script.
Changes:
- Add
reviews/target_audience.mdwith JMD reviewer-profile diversity expectations, a named candidate reviewer pool, a COI screen, a lander-style outreach contact list, and draft first-contact talking points. - Commit Edison trajectory artifacts (
.md+.json) for the target-audience and outreach-topicsLITERATURE_HIGHruns. - Add
scripts/edison/submit_outreach_topics.pyto reproduce the outreach-topics Edison task and persist results.
Reviewed changes
Copilot reviewed 6 out of 6 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
| scripts/edison/submit_outreach_topics.py | Submits and fetches an Edison LITERATURE_HIGH outreach-topics query; writes submitted task_id + md/json artifacts. |
| reviews/target_audience.md | JMD/SMS audience analysis, candidate reviewer pool, outreach contact shortlist, and draft talking points. |
| edison-trajectories/2026-05-09-target-audience-9cc7db18.md | Verbatim Edison formatted answer for the JMD reviewer/AE profiling query. |
| edison-trajectories/2026-05-09-target-audience-9cc7db18.json | Structured Edison task payload for the target-audience query. |
| edison-trajectories/2026-05-12-outreach-topics-f18aca01-00bb-4ca7-a8e9-f6312dfaaff7.md | Verbatim Edison formatted answer for outreach talking points / ask-routing. |
| edison-trajectories/2026-05-12-outreach-topics-f18aca01-00bb-4ca7-a8e9-f6312dfaaff7.json | Structured Edison task payload for the outreach-topics query. |
| - **Manufacturing:** PETG struts + TPU 95A tendons on a Bambu Lab H2D IDEX | ||
| printer (per the lab's hardware standard -- see the repo's hardware memory). |
| > | ||
| > **Primary venue:** ASME Journal of Mechanical Design (JMD). | ||
| > **Backup venue:** Smart Materials and Structures (SMS, IOP). | ||
| > See `manuscript/README.md` (PR #20) for the venue selection rationale. |
| - `edison-trajectories/2026-05-08-sim-survey-782657e0.md` | ||
| (branch `copilot/explore-simulations-for-tensegrity`). |
| # Persist the task id immediately so we can fetch later if needed. | ||
| submitted_path = OUT_DIR / "outreach-topics-SUBMITTED.json" | ||
| submitted_path.write_text( | ||
| json.dumps({"task_id": task_id, "name": "LITERATURE_HIGH"}, indent=2) | ||
| + "\n" | ||
| ) | ||
| print(f"wrote {submitted_path}") |
| # Try to wait in-session. | ||
| print("Waiting for task to complete (this may take several minutes) ...") | ||
| results = client.get_task(task_id, verbose=True) | ||
| if not isinstance(results, list): | ||
| results = [results] | ||
| t = results[0] | ||
| print(f"status = {t.status}") | ||
|
|
||
| md_path = OUT_DIR / f"2026-05-12-outreach-topics-{task_id}.md" | ||
| json_path = OUT_DIR / f"2026-05-12-outreach-topics-{task_id}.json" | ||
| md_path.write_text(getattr(t, "formatted_answer", "") or "") | ||
| json_path.write_text( | ||
| json.dumps(json.loads(t.model_dump_json()), indent=2) + "\n" | ||
| ) | ||
| print(f"wrote {md_path}") | ||
| print(f"wrote {json_path}") | ||
| return 0 |
|
@me-madsen @achris0520 @ctrhjk could one of you work on reaching out to the contacts from the following sections? Noting there's some overlap, so making sure we don't email the same person twice. Here's an example of previously when we've been reaching out for another project, modified to match this project more closely: We can use this for the picture: 
Also for the video (a trimmed version of Signal1 from #71 (comment)): tensegrity-drop.mp4 |
I'm on it 🫡 |
|
Thank you! Please make sure to cc me on the emails |
|
@me-madsen thinking about it some more. up to you whether you cc just me or not. Whatever you think is best. When you send the first few emails, could you post back here mentioning who you reached out to? |
I didn't see this until after I sent the first 3, but I will likely adjust my cc list for future emails. For documentation, I just reached out to: Julian J. Rimoli |
|
Additional emails have been sent to the following, with only @sgbaird ccd: |
|
This is a paper from another research group studying tensegrity crutches. 'Tensegrity crutches with compliance from a pre-stressed self-tensile module improve ground reaction force profiles, speed, effort, comfort, and perceived stability' First author: Jingxian Gu (jgu@unomaha.edu) |
|
Thanks, Jinkwan! Could you reach out to the two of them to see if they'd be up for chatting about the paper and providing some feedback to us, then report back here when you've sent it? (and the text of what you sent) (please cc me) |
|
@sgbaird Here are the text of what I sent Dear Dr. Dobromir Dotov and Jingxian Gu, My name is Jinkwan Han, and I am an undergraduate research assistant in Dr. Jeff Hill and Dr. Sterling Baird's lab at Brigham Young University. We recently saw your work at the biomechanics conference and read your manuscript, "Tensegrity crutches with compliance from a pre-stressed self-tensile module improve ground reaction force profiles, speed, effort, comfort, and perceived stability", and found it very relevant to our current research on 'optimizing the mechanical properties of multi-material, tensegrity-like structures'. We are studying similar topics and found your work especially helpful in understanding the experimental design and interpretation of the results. Attached is: I was wondering if either of you would be willing to meet with our team for a short virtual conversation. We would love the opportunity to ask a few questions about your study and get your feedback as we continue our own research. If you are available, we'd be happy to work around your schedule. Thank you for your time and for your contributions to this area of research. We appreciate your consideration and hope to hear from you. Best regards, Jinkwan Han 
Video.Project.21.mp4 |
Dr. Filipe A. Santos and Dr. Andrea Micheletti have individually replied back and agreed to meet. We're in process of arranging a time to meet with them virtually and to accommodate the significant time differences between our respective locations. Dr. Santos expressed specific interest in particular aspects of our work: "I was particularly intrigued by your ability to fabricate multi-material tensegrity-inspired structures in a single FDM printing process, and I look forward to learning more about your manufacturing approach and discussing your performance assessment and optimization strategy." Will update once I gain a response back from them. |
|
Additional emails have been sent to: |
|
Notes from meeting with Dr. Haris Moazam Sheikh: We reviewed the search space we have running for our experiment on current variables as noted in #85 (comment) Dr. Sheikh reflected that the search space appeared achievable and recommended keeping to no more than 2 objectives, and said that having 6 search variables is good, and that we could have up to 20 or 25. He recommended we look at measuring/quantifying white noise so that it could be fed into the BO model (i.e. quantifying and recording print defects, differences in the CAD model verses the print model, etc.). Additional notes from @achris0520 below. |
|
Notes from Our Meeting With Dr. Haris Moazam Sheikh: He recommends using no more than 2 objectives. If you have to have more, just treat one as a constraint. Recommendations for Learning About Bayesian Optimization: Design Problems: Handling the noise from models with defects: |
|
Meeting notes with Dr. Filipe A. Santos (transcription & recording available but not uploaded): For context, Dr. Santos has worked with 3D printed tensegrity structures, where his were assembled and tensioned post-printing. Dr. Santos noted that our process is quite interesting as it seems a very novel idea to try to print tensegrity related structures as one assembled unit. His feedback is best summarized with his statement: "You can say that it's a tensegrity-inspired [structure], and then you can try to optimize the cross-section or the shape of the tie or the shape of the truss, but I think that you are not looking into the most important aspect of these particular structures, which is pre-stress." Dr. Santos expressed concerns about the following:
Dr. Santos suggested the following:
We also discussed our recent issues with printing the tendons of our structures as documented in this comment and its associated pr #35 (comment). Dr. Santos expressed that it was impressive we've been able to print our tendons at such a steep incline at all, and that these issues introduced variability in our experiments. Dr. Santos expressed that he was happy for us to meet again or communicate in the future via email. |
|
Notes from meeting with Dr. Liwei Wang (7-28-2026). Video recording available but not posted. I've been using this slideshow with recent meetings to show our work and process: https://byu.box.com/s/5k9os5cijnm2dbn0mrtngzglph6ce1do Dr. Wang thought the BO loop we have set up or thought out was a good and standard application of BO. He really liked that we're using and applying real data to the loop and not just simulation data.. Dr. Wang suggested the following, some of which are items we have planned:
Other notes:
|
|
Notes from meeting with Dr. Dobromir Dotov and Jingxian Gu. @achris0520 and @ctrhjk feel free to edit this with further notes or add your own notes in a separate comment. This meeting was focused around crutch tip application of the tensegrity-inspired specimens that we're optimizing, as noted here #18 and in @ctrhjk's abstract for TMS 2027. For context, Dr. Dotov and Jingxian published this paper where they explored using true tensegrity devices for crutch tips and found them to have a positive impact. This is a link to their paper: https://doi.org/10.48550/arXiv.2605.02710 Their research is very recent, having just emerged this year. They expressed some of the following sentiments about tensegrity crutch tips:
They are interested in collaborating with our lab as they would like to explore application of 3D printed tensegrity devices, particularly to walking stick tips as they're not confident that the 3D printed structures could be stiff enough or withstand the kind of cyclic loading that crutch tips would undergo with regular use. Their background is in biomechanics, and they expressed that the help of mechanical engineers would be quite helpful. This is a proposition to discuss with @sgbaird and @Jeffrayhill1. Additional notes on this front:
|
The camera slides had turned into an apologia for what 960 fps cannot resolve — a question the audience never asked. Messages are now what the video actually shows (spring-back, intact specimen, rebound vs. force), with the frame-rate arithmetic moved to presenter-only notes for Q&A. Also adds previews of @me-madsen's Workflow Discussion deck (Box, #41). Co-authored-by: Sterling G. Baird <45469701+sgbaird@users.noreply.github.com>
Identifies likely Associate Editors, reviewer-profile diversity, and named candidate reviewers for the PR #20 manuscript (primary: ASME JMD; backup: SMS), plus a separate cluster of named contacts for "lander"-style / drag-free egg-drop outreach (PR #47), plus drafted first-contact talking points for outreach emails — all grounded in dedicated Edison
LITERATURE_HIGHqueries whose.md+.jsonartifacts are committed alongside this analysis.Changes
edison-trajectories/2026-05-09-target-audience-9cc7db18.{md,json}(new) — verbatimformatted_answer(~40 KB, 25-entry numbered reference list with DOIs and citation counts) plus the structuredmodel_dump_jsonpayload from EdisonLITERATURE_HIGHtask9cc7db18-10b5-457e-9b7c-9a3ecb2b9f14. Per-publication detail (DOIs, citation counts, page-level evidence anchors) is preserved in the JSON for full reproducibility.edison-trajectories/2026-05-12-outreach-topics-f18aca01-00bb-4ca7-a8e9-f6312dfaaff7.{md,json}(new) — verbatimformatted_answer(~50 KB md, 89 KB JSON, status: success) from EdisonLITERATURE_HIGHtaskf18aca01-00bb-4ca7-a8e9-f6312dfaaff7, grounded in cross-PR context (PRs Populate ASME Journal of Mechanical Design manuscript draft from repo context, with todonotes toggle, consolidated bibliography, Edison mock peer review, first-pass blocker fixes, cross-PR planned-methods refresh, a master synthesized bibliography, rea... #20, Document Lansmont M23 drop tower and Polytec QTec laser vibrometer #28, Extensively explore and validate the joint design for PETG + TPU prints to mimic tensegrity #38/Edison-driven survey of funding venues for a larger-scale tensegrity / BO grant #43, Explore egg-drop demo via Edison LITERATURE_HIGH query (+ drag-free baseline / V·m benchmark, Brown-lab BEAR baseline, and SUPERball v2 actuator follow-ups) #47). Returns 7 archetype clusters in (a), tech-transfer angles in (b), 8 gotchas in (c), 8 collaborative mechanisms in (d), per-contact ask routing for all 12 named lander-pool contacts in (e), and an 8-item "stretch" section (digital-twin round-robin, materials-genome integration, citizen-science egg-drop kit, prestress-monitoring IoT, PolyFEM topology-opt collab, ASTM F42/D20 standards pathway, FTO patent landscape scan, high-speed-video + accelerometer sharing consortium).scripts/edison/submit_outreach_topics.py(new) — reproducible submission script for the outreach-topics Edison query, with the full grounded prompt embedded.reviews/target_audience.md(new)reviews/mock_reviews.md.egg-drop-tensegrity-1b90208d.mdandegg-drop-followup-f41b7034.md. Suggested first-contact set of 3: Julian J. Rimoli (Georgia Tech — On the impact tolerance of tensegrity-based planetary landers, AIAA SciTech 2016), Adrian K. Agogino (NASA Ames — NIAC SUPERball Phase 2 final report 2018, the only published egg-payload tensegrity drop with quantitative deceleration data), Mark W. Mueller (UC Berkeley HiPeRLab — Zha et al., IEEE/ASME Trans. Mechatronics 2024 collision-resilient icosahedron). Tier-2 (4–10): Vytas SunSpiral, Andrew Zhang + Brian Cera (UC Berkeley, Agogino lineage), Massimo Vespignani (SUPERball v2), Skelton/Sultan, Jamshid Bayandor (Virginia Tech CRASH Lab — TANDEM 2017), Jing Zhang et al. (HIT — Aerospace 2025), Madhumati Anand (biodegradable tensegrity, 75 m drops). Notes that this cluster is essentially disjoint from the crutch-tip / assistive-device pool (only Rimoli + Pajunen are shared with §3b/§3d), so a space-systems / aerospace-mechanisms reviewer should be expected if the egg-drop demo is folded into the manuscript.f18aca01for any actual email body..mdfiles on PR branches plus the consolidatedmanuscript/references.bib(PR Populate ASME Journal of Mechanical Design manuscript draft from repo context, with todonotes toggle, consolidated bibliography, Edison mock peer review, first-pass blocker fixes, cross-PR planned-methods refresh, a master synthesized bibliography, rea... #20). Surfaces that the current corpus is materials/AM-heavy (Polymers, Mater. & Des., Addit. Manuf.) and light on core ASME design-theory venues (one J. Mech. Robotics hit, one SMS hit). Action item: thicken with a targeted JMD/JAM/JMR/S&MO pass before submission.9cc7db18as the primary source for §3a–§3e, the two PR Explore egg-drop demo via Edison LITERATURE_HIGH query (+ drag-free baseline / V·m benchmark, Brown-lab BEAR baseline, and SUPERball v2 actuator follow-ups) #47 egg-drop trajectories (1b90208d+f41b7034) as the source for §3d′, the new outreach-topics trajectoryf18aca01as the source for §3d″, plus all 19 prior trajectory files andreviews/mock_reviews.md.