About the evidence
This is a reference book, not a promotional one. Its usefulness depends on a reader being able to see where a claim came from, how strong it is and what it does not cover. This page explains the evidence grades, the citation system, the way numbers are presented, and the known limits of the record.
What the book is
The book analyses the 2027 University Rover Challenge for a university team building a Mars rover for the first time. It covers the competition rules and scoring, the teams and their machines, twelve subsystems, the cross-cutting problems of mass, power, interfaces and failure, the 2027 mission backfit, and a build plan. The print edition is the same text in a fixed A4 layout; this edition adds the full source list, the asset record and the interactive tables.
The unit of evidence is a source, not a paragraph. Every non-trivial factual claim carries a source identifier. Where the record does not contain an answer, the text says so rather than filling the space with a guess.
Evidence grades
Four grades appear in the deeper pages and in the data tables. They are visible as small pills.
| Grade | What it means | How it is used |
|---|---|---|
| Primary | Stated in a primary source: the team’s own report, site, code or video, an official URC document, official results, or a manufacturer’s datasheet. | The strongest grade. It still depends on the source being accurate and current. |
| Reported | A credible secondary source: a news article, a university release, a sponsor case study or a competition write-up. | Useful for context and for facts a team does not publish itself. It is one step removed from the team. |
| Inference | A reasoned deduction from images, video, partial data or comparable systems. | The text says what the inference rests on. An inference is not a measurement. |
| Not published | No public figure exists, and the text says where it was looked for. | The honest answer. It becomes a measurement task or an open question, not a silent omission. |
A statement without a grade is either a direct quotation, a description of the book’s own analysis, or a definition. A grade is not a score: a well-sourced claim can still be wrong if the source is wrong, and an inference can be more useful than a weak primary figure.
How a claim is worded
The reading text rarely prints a grade beside a sentence. The wording carries it instead, and the wording is deliberate.
| Reading | What it tells a reader |
|---|---|
| A source identifier with nothing beside it | A primary source states it: a team document, an official URC document, a datasheet or an official score sheet |
| “the team reports”, “according to the team” | A credible secondary account, one step removed from the team’s own record |
| “estimated”, “by inference”, “target derived rather than published” | The book’s own reasoning from stated parameters, images or comparable systems |
| “not published”, “not stated in the rules”, “unmeasured” | The record has no answer, and the text says where it was looked for |
A value taken directly from a source always carries that source’s identifier. A value that the book derived is stated with its inputs, so the arithmetic can be repeated and argued with.
How sources are cited
Every source has a stable identifier of the form S-TOPIC-SHORT, for example S-URC27-RULES for the 2027 rulebook or S-MST-2026-SAR for Missouri S&T’s 2026 review film. The identifier is the same everywhere it appears, so a claim can be traced from a page, to the bibliography, to the original document.
In the text, an identifier is a small link. It opens the entry in the bibliography, which gives the title, author or publisher, date and an original link where one exists. A page also ends with a list of the sources it used, so the evidence for a single page can be read without leaving it.
Rule and Q&A references work the same way. A clause such as <a class="clause" href="../competition/rules.html#clause-2a">§2.a.ii</a> links to the rules page; a Q&A identifier such as RQ11 links to the Q&A page. Requirement identifiers such as <a class="ref" href="../competition/requirements.html#req-req-mob-004">REQ-MOB-004</a> link to the requirements table.
The bibliography is not a reading list. It includes vendor datasheets, team code, archived pages and searches that found nothing. A source can be cited because it is the best available evidence for a fact, not because it endorses the conclusion.
How to read the numbers
Five questions settle most disagreements about a number:
- What is the sample? A score statistic may cover 162 finals team-years from 2022 to 2026, 36 teams in one year, or 17 rover configurations from eight teams. The page states the sample next to the number.
- What is the period? Rules, scores and designs change. A 2024 figure is labelled 2024, and a claim about 2027 is labelled 2027.
- Who owns the value? A number belongs to the analysis that derives it, not to the page that quotes it. The key numbers page gives the owner and the source for each headline value.
- Is it measured, published or inferred? A team-published mass is not a judge-weighed mass. A CAD sum is not a scale reading. The basis is stated.
- What is the unit and the rounding?
12 %has a space, ranges are written 2022-2026, and a figure is not rounded from47.3 N·mto47 N·m. A value that is an estimate says so.
Two places need special care. First, the 2027 task weights for Astrobiology, Delivery and Equipment Servicing are not published, and the Autonomy Mission is the only one whose sub-task points are given S-URC27-RULES §1.e.iv. Second, the 2027 review and Science Plan specifications were not published when this edition was prepared, so the 2026 documents are used only where they are labelled as such S-URC27-SAR S-URC27-SCIENCEPLAN.
What the evidence covers well
The rules, the official scores and the team-published material are strong. The 2027 rulebook, the 2022 to 2026 score sheets and the 2027 Q&A snapshot are kept as archived copies. The 2022 to 2026 results are complete enough to support distributions, zero rates and variance decomposition. The top teams publish enough video, code and documentation to reconstruct their architecture at the subsystem level.
The record is weaker in four places:
- Mass and centre of gravity. Most teams publish no mass, and fewer publish a centre of gravity. The mass model rests on 17 configurations, 13 of them published masses, plus arithmetic from stated reductions and CAD sums. It is a good model of the field, not a measurement of any one rover.
- Task-level forces. The 2027 rules describe the equipment servicing tasks but do not publish the forces, travels or torques. Those numbers have to be measured on a mock. The book gives design targets and says they are targets.
- Delivery and drones. The delivery study is a synthesis from the rules, the Q&A and team evidence rather than a dedicated study, and the page says so.
- The qualification funnel. Nineteen of the 68 census teams have no profile because they publish almost nothing. They are listed in the census and represented by their score row. Two further teams that registered but never reached the finals sheet have profiles because their public engineering record is unusually good.
A page-by-page account of what is missing, measured or inferred is on evidence coverage.
What the book does not do
The book does not design the rover. It gives the requirements, the evidence, the options and the trades, and it recommends a first-build shortlist. The detail design, the test campaign and the build are the team’s work.
It does not reason about the team’s budget, funding or ability to pay. Cost appears only as a relative property in a comparison table, from $ to $$$$$, and as sourced list prices with a date. The competition’s own $26,000 fielded-equipment cap is recorded as a competition rule, not as a constraint on the analysis S-URC27-RULES §3.c.
It does not claim that a choice which worked for a top team will work for a first build. The dossiers show what teams did and what it cost them. The subsystem pages separate the evidence from the recommendation.
Corrections and reuse
The original analysis and the charts made for this book may be reused with attribution to the Robotics Club at Mississippi State University. Quotations, photographs, team documents and vendor images remain the property of their owners and are credited with a link to the original. Where a licence is known it is shown; where it is not, the credit names the author and the source. The credits page states the terms in full.
Corrections are welcome. A correction should name the page, the claim, the source and the replacement value. The standard is the same as the one used here: a number needs an owner, a sample and a date.
The one rule
A claim may be wrong, but it may not be unsourced. Where the record has no answer, the book says so.
Sources for this page
- S-URC27-RULESUniversity Rover Challenge 2027 – Requirements and GuidelinesOriginal Archived copy
- S-URC27-SARURC System Acceptance Review page, states "URC2027 SAR Requirements will be posted in early 2027!"Original
- S-URC27-SCIENCEPLANURC Science Plan page, states "URC2027 Details will be posted in early 2027!"Original