1. The short version / what to know first / where this fits in the series

• Surgical robots are tele-operated. The FDA says that such a device cannot perform surgery without direct human control, so it is not actually a robot Sourced[Ref. 1]. The surgeon moves hand controllers at a console, and the motion is carried to the arms and instruments on a cart beside the patient.
• “Approval” and “clearance” are different words, and the scope is set per type of surgery. FDA documents describe the da Vinci 5 for urologic, general laparoscopic, gynecologic laparoscopic and general thoracoscopic procedures in adults (documents of 14 March 2024 and 16 July 2025) Sourced[Ref. 6, 7]. Medtronic’s Hugo was cleared on 3 December 2025 for minimally invasive urologic procedures in adults Sourced[Ref. 8].
• Machines that perform surgery autonomously are at the research stage. A 2022 study had a robot suture intestine autonomously in pigs, but within this article’s sources no marketed device doing this in human surgery could be confirmed Sourced[Ref. 21]. In September 2026 the FDA issued a draft guidance on premarket submissions for such devices; it is a draft and not yet in force Sourced[Ref. 5].

2. What a surgical robot is: the robot does not do the surgery

Start with laparoscopic surgery. A few small holes are made in the abdomen. A camera (endoscope) goes in through one, long thin instruments go in through the others, and the surgeon moves the instruments while watching a screen (explanation). A surgical robot helps with exactly this task of “watching a screen and moving long thin instruments”.

The U.S. Food and Drug Administration (FDA) calls these devices robotically-assisted surgical (RAS) devices and describes them as devices that let the surgeon use computer and software technology to control instruments. The parts are a console, a patient-side cart with mechanical arms, and a separate cart with supporting hardware. On the same page the FDA says the device cannot perform surgery without direct human control, so it is not actually a robot Sourced[Ref. 1].

Map the figure onto the three elements of a robot. Sensing is the endoscope camera, whose video reaches the surgeon’s eyes (force sensors exist only in some instruments). Thinking is mainly the surgeon. The control computer adjusts the motion ratio, filters tremor and monitors safety, but it does not decide how the operation proceeds. Moving is done by the motors in the arms and instruments (explanation).

3. How hand motion reaches the instrument: pivot, wrist and scaling

(1) The problem: the pivot reverses the motion

The instrument in conventional laparoscopic surgery is a straight stick. It pivots about the small hole in the abdominal wall, so when the hand moves left, the tip inside the body moves right. Think of a seesaw: the far side of the pivot moves the opposite way. (Limits of the analogy: a real stick also slides a little at the hole.) The stick also has no joint at its end, so the tip cannot be bent to change its direction (explanation).

Many surgical robots soften this with two devices. One is an instrument that bends like a wrist. Intuitive says most of its instruments have wristed joints for natural dexterity, with tips such as forceps, scissors and electrocautery tools shaped for each procedure Sourced[Ref. 17]. Hugo’s 510(k) summary also lists nine wristed instruments (needle drivers, graspers, shears and others) Sourced[Ref. 8]. The other is software that corrects direction, so the direction of the surgeon’s hand and the direction of the tip agree (panel 2 is a concept diagram of the idea).

(2) Scaling the motion and reducing tremor

Hand motion at the console is not copied one-to-one. A ratio can scale it down. Intuitive says its technology translates the surgeon’s hand movements into precise real-time micro-movements of the instruments, uses motion scaling to support precision in delicate tasks, and filters the tremor of the hands Sourced[Ref. 17].

Check it with numbers. Assume a 3:1 ratio. When the hand moves 30 mm, the instrument tip moves 30 / 3 = 10 mm Our calculation. A 1.5 mm hand tremor becomes 1.5 / 3 = 0.5 mm at the tip Our calculation. (The ratio is an assumed value for illustration; actual ratios were not checked.) Scaling shrinks the tremor by the same factor. A filter is a separate process that removes tremor on top of this.

(3) Touch has long been a weak point

A known weakness of tele-operation is that no sense of force comes back to the surgeon’s hands. Intuitive says that the da Vinci 5, which the FDA cleared in 2024, has instruments that let surgeons sense and measure the force exerted on tissue (force feedback) Sourced[Ref. 17]. According to K232610, a sensor on the instrument detects directional forces and sends them back through the hand controls on the console, and six force-feedback instruments, including needle drivers and graspers, were added Sourced[Ref. 6].

The same document states that the force-feedback needle driver is contraindicated (must not be used) in hysterectomy and myomectomy, and recommends non-force-feedback needle drivers for suturing in those procedures Sourced[Ref. 6]. A new function therefore has a scope that the documents define in detail, procedure by procedure. Intuitive also states that the European certification of the da Vinci 5 (July 2025) excludes force feedback Sourced[Ref. 17].

4. Inside the instruments, and materials: cables, fatigue, cleaning and sterilization

It is often said that much of a surgical robot’s performance is set at the thin instrument wrist at the end of the arm. Let us look inside the instrument as a materials engineer.

Cables move the wrist

The shaft of an instrument is narrow, so motors cannot sit near the wrist. Many surgical robots therefore pull the wrist and jaws with cables from a drive unit at the base of the shaft. It resembles a puppet on strings. (Limits of the analogy: real cables run over several pulleys, and several are pulled in coordination at once.) Hugo’s 510(k) summary calls the system “cable-connected” capital equipment, but that describes how the main equipment is wired, not how the instruments are driven Sourced[Ref. 8]. The cable material and construction inside the instruments could not be confirmed from official sources for any specific model.

The general problems of cable-driven mechanisms have been studied. A 2017 University of Washington PhD thesis, using the RAVEN II surgical robot as its basis, reports that cable-driven mechanisms face control challenges from the hysteretic stretch of the cable and the dynamic friction of the guide pulleys, and it investigates the cable’s fatigue life and presents a prediction model Sourced[Ref. 23]. Hysteresis means that the stretch differs between loading and unloading at the same force. These are results for a research robot, not lifetimes of commercial systems.

Wash, sterilize, use again

Reusable instruments must be cleaned, inspected and sterilized for each surgery. Hugo’s 510(k) summary lists non-clinical tests that include sterility testing, shelf-life testing, reprocessing (cleaning and sterilization) validation, and biocompatibility and pyrogenicity testing Sourced[Ref. 8]. The same summary says the drapes and instrument tip covers are single-use (packaged sterile) and that the reusable instruments are not shipped sterile (the hospital processes them) Sourced[Ref. 8]. The da Vinci 5 endoscope tray is stainless steel, and its use is described for several sterilizers and cycles such as STERRAD and STERIS V-PRO Sourced[Ref. 6].

5. FDA “clearance” and “approval”: three routes

In the United States the FDA reviews medical devices before they are marketed. The route depends on the risk and novelty of the device. News often says “FDA approved”, but the word depends on the route.

• 510(k) (premarket notification): a route in which the device is shown to be “substantially equivalent” to an earlier predicate device already on the market. The FDA’s result is called “clearance”. Substantial equivalence does not require the devices to be identical Sourced[Ref. 2].
• De Novo: a route for a new type of device with no predicate. When granted, a new classification regulation is created and the device can serve as a predicate for later 510(k)s Sourced[Ref. 3].
• PMA (premarket approval): the strictest review, required for Class III devices and some high-risk new devices. The result is “approval” Sourced[Ref. 4].

Within what could be confirmed here, the surgical robot platforms reached the market through 510(k) or De Novo. A search of FDA’s data returns more than 100 510(k) records with “da Vinci” in the name [Ref. 16], and Intuitive says that the first da Vinci was cleared by the FDA in 2000 for general laparoscopic surgery Sourced[Ref. 17]. The product code is NAY for da Vinci (Class II, 21 CFR 876.1500) [Ref. 11] and SCV for Hugo and Versius (Class II, 21 CFR 878.4964) [Ref. 10]. The definition of SCV reads, in substance, a software-controlled electromechanical system that lets a qualified user perform surgical techniques during minimally invasive procedures Sourced[Ref. 10].

A chain: a new route becomes the benchmark for the next device

De Novo and 510(k) are linked. CMR Surgical’s Versius was granted De Novo (DEN230078) on 11 October 2024 (received 21 November 2023), creating a new classification (product code SCV) Sourced[Ref. 9]. Medtronic’s Hugo was then cleared in a 510(k) (K250725) on 3 December 2025 with Versius as its predicate. Hugo’s summary records the predicate’s indication as cholecystectomy in adults aged 22 or older who are eligible for minimal-access soft-tissue surgery Sourced[Ref. 8].

6. Reading the scope by type of surgery: the same machine differs by country and model

The “Indications for Use” section of an FDA document is the text that fixes what the device is for, in whom, and in which procedures. The table and figure below line up what the documents checked here state.

Checked 25 October 2026. The table summarizes what the documents checked here state. Read the full documents (and the FDA databases) for the actual indications.

Surgical robots are not one kind of machine. Tele-operated systems for soft tissue inside the abdomen (da Vinci, Hugo, Versius), systems for cutting bone in orthopedics (Mako), and systems that advance a thin tube deep into the lung (Ion) are authorized under different product codes and different indications (explanation). Lumping them as “surgical robots” hides these differences.

Even the same da Vinci 5 has a different scope by country

Intuitive’s 10-K describes the scope of da Vinci 5 authorizations by region as follows (the company’s description) Sourced[Ref. 17].

This summarizes the company’s own description; the regulators’ original texts (Korea, Japan, EU) were not checked. The U.S. scope is consistent with FDA documents K232610 and K251202.

Japan’s hinotori (Medicaroid, a company funded by Kawasaki Heavy Industries and Sysmex) was announced to have obtained regulatory approval in Japan in August 2020 Sourced[Ref. 19]. That announcement does not state which surgeries the approval covers, and the scope could not be confirmed from regulator documents within this article’s research.

7. Levels of autonomy, and where research stands

There is a proposal to think about “how much does the robot act by itself” in levels. Yang et al. (2017) divided the autonomy of medical robots into six levels, from Level 0 (no autonomy) to Level 5 (full autonomy) Sourced[Ref. 20]. It resembles the “levels” of self-driving cars, but because medicine involves a wide range of settings and risks, the paper says the same level can mean different things in different contexts.

In this framework a tele-operated system with scaling, the basic form of today’s surgical robots, is Level 0. For Level 2 the paper’s example is the surgeon indicating where a suture should be placed, the robot performing it, and the surgeon monitoring. Level 4 is the analogy of a “robotic resident” making medical decisions under supervision. For Level 5, a “robotic surgeon”, the paper wrote in 2017 that it is “in the realm of science fiction” Sourced[Ref. 20].

A research example: STAR (suturing intestine in pigs)

Saeidi et al. (2022) reported a method in which a robot autonomously performs intestinal anastomosis (joining two ends of bowel by suturing) laparoscopically. The operator selects among autonomously generated surgical plans, and the robot executes a wide range of tasks on its own. They evaluated it in phantoms and in live pig small intestine with a one-week survival period Sourced[Ref. 21]. In phantoms, the authors report that their system outperformed expert surgeons’ manual technique and robot-assisted technique in consistency and accuracy, and say the same held in vivo. This is, however, a study in pigs; it does not show use in human surgery or marketing. In terms of Yang et al.’s definitions, “the robot makes the plan and the human selects it” reads close to a Level 3 description (inference).

FDA draft guidance (September 2026)

In September 2026 the FDA issued a draft guidance titled “Robotically-Assisted Surgical Devices – Premarket Submissions”. It gives recommendations for premarket submissions on non-clinical testing, clinical data and labeling, and it defines RAS devices as teleoperated, software-controlled systems. The page states that it is not for implementation and contains non-binding recommendations; comments are due 24 November 2026 Sourced[Ref. 5]. Whether the draft will be finalized, and how it may change, cannot be confirmed Not yet confirmed.

8. How to read claims about effectiveness and safety (without concluding)

“Is it good for patients?” is a natural question, but this article does not judge effectiveness or safety. Instead, here are three cautions for reading documents.

1. Look at the study design. The clinical study in Hugo’s 510(k) summary (Expand URO) was a prospective, multicenter, non-randomized, single-arm study of 137 people (29 cystectomy, 53 nephrectomy, 55 prostatectomy) at six U.S. sites, with data collected through 30 days after the procedure Sourced[Ref. 8]. 29 + 53 + 55 = 137 Our calculation. A single-arm study is not designed to compare directly with other methods. Medtronic announced that the study met its primary safety and effectiveness endpoints (a company claim) [Ref. 18].
2. A company’s numbers are the company’s numbers. Intuitive reports about 11,106 da Vinci systems installed at the end of 2025, 1,721 new placements in 2025 and 1,231 da Vinci 5 systems installed Sourced[Ref. 17]. The da Vinci 5 is 1,231 / 11,106 = about 11% of the installed base Our calculation. These figures show the scale of sales, not how good the treatment results are.
3. Read what the FDA has not done. The FDA page says it has not granted marketing authorization to RAS devices for cancer prevention or treatment, or for mastectomy Sourced[Ref. 1]. Absence of authorization does not mean “dangerous” or “ineffective”, but it does mean that outside the authorized scope the FDA has not evaluated the use (explanation).

9. Where things stand (checked 25 October 2026)

This field changes quickly, so here is what could be confirmed from primary sources, with check dates. Company claims are written as company claims.

Checked 25 October 2026. Company plans and targets are separate from achieved facts.

10. How to read the yardsticks and standards

The quality of a surgical robot cannot be measured by one number. Machine performance such as positional accuracy, motion delay and instrument life can be measured in a laboratory. But “how is it for patients” is judged by clinical studies and evaluation per type of surgery, and “may it be marketed” is decided by regulatory review (510(k), De Novo, PMA). These yardsticks measure different things. The table below shows what each measures and does not measure, as far as could be confirmed.

The arrangement of the table is Our calculation (organization). Numbers and titles of standards and rules were checked on the official FDA and IEC pages (25 October 2026). The IEC text is paid and was not read, so only the title and scope are given. No ASTM International standard or website was consulted.

Cautions when reading (for materials engineers)

• “Machine performance”, “clinical results” and “market authorization” are different yardsticks. High positional accuracy is not the same as good surgical results. FDA clearance indicates substantial equivalence to a predicate (or a new classification), not superior outcomes.
• Separate company announcements from regulator documents. Installed base, procedure counts and “met the primary endpoints” are company claims. The FDA’s 510(k) summary is a closer-to-primary source that shows the design and scope of tests.
• A standard is a “reference frame”, not law itself. IEC 80601-2-77 is one standard used to demonstrate conformity within national and regional regulations (inference). Its text was not read here, so its requirements are not described.
• Plans and achievements differ. The draft guidance and Medtronic’s targets for added indications are not yet confirmed.

11. Glossary

12. References

1. FDA “Computer-Assisted Surgical Systems” (FDA web page, content shown as current to 24 September 2026; checked 25 October 2026) — fda.gov/medical-devices/surgery-devices/computer-assisted-surgical-systems
2. FDA “Premarket Notification 510(k)” (FDA web page, current to 22 August 2024; checked 25 October 2026) — fda.gov/medical-devices/premarket-submissions-selecting-and-preparing-correct-submission/p
3. FDA “De Novo Classification Request” (FDA web page, current to 30 September 2025; checked 25 October 2026) — fda.gov/medical-devices/premarket-submissions-selecting-and-preparing-correct-submission/d
4. FDA “Premarket Approval (PMA)” (FDA web page, current to 16 May 2019; checked 25 October 2026) — fda.gov/medical-devices/premarket-submissions-selecting-and-preparing-correct-submission/p
5. FDA “Robotically-Assisted Surgical Devices – Premarket Submissions” (draft guidance, September 2026; comments due 24 November 2026, docket FDA-2026-N-9505; checked 25 October 2026) — fda.gov/regulatory-information/search-fda-guidance-documents/robotically-assisted-surgical
6. FDA 510(k) K232610, da Vinci Surgical System Model IS5000 (decision date 14 March 2024; indications, force-feedback instruments and predicates; checked 25 October 2026) — accessdata.fda.gov/cdrh_docs/pdf23/K232610.pdf
7. FDA 510(k) K251202, da Vinci Surgical System (IS5000) (decision date 16 July 2025; indication text; checked 25 October 2026) — accessdata.fda.gov/cdrh_docs/pdf25/K251202.pdf
8. FDA 510(k) K250725, Hugo RAS System (Covidien LLC, decision date 3 December 2025; indications, predicate, performance testing, Expand URO study summary; checked 25 October 2026) — accessdata.fda.gov/cdrh_docs/pdf25/K250725.pdf
9. FDA De Novo DEN230078, Versius Surgical System (CMR Surgical; received 21 November 2023, granted 11 October 2024; checked 25 October 2026) — accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/denovo.cfm?id=DEN230078
10. FDA (openFDA) Product code SCV, “Modular Electromechanical Surgical System” (21 CFR 878.4964, Class II; checked 25 October 2026) — api.fda.gov/device/classification.json?search=product_code:SCV
11. FDA (openFDA) Product code NAY, “System, Surgical, Computer Controlled Instrument” (21 CFR 876.1500, Class II; checked 25 October 2026) — api.fda.gov/device/classification.json?search=product_code:NAY
12. FDA (openFDA) Product code OLO, “Orthopedic Stereotaxic Instrument” (21 CFR 882.4560, Class II; checked 25 October 2026) — api.fda.gov/device/classification.json?search=product_code:OLO
13. FDA (openFDA) Product code EOQ, “Bronchoscope (Flexible Or Rigid)” (21 CFR 874.4680, Class II; checked 25 October 2026) — api.fda.gov/device/classification.json?search=product_code:EOQ
14. FDA (openFDA) 510(k) records search: Mako Surgical Total Knee, Total Hip and Shoulder applications, for example K260222 (25 February 2026, product code OLO; checked 25 October 2026) — api.fda.gov/device/510k.json?search=device_name:%22mako%22&limit=100&sort=decision_date:de
15. FDA (openFDA) 510(k) records search: Intuitive Ion Endoluminal System, for example K182188 (14 February 2019, product code EOQ; checked 25 October 2026) — api.fda.gov/device/510k.json?search=device_name:%22ion%22+AND+applicant:%22intuitive%22&li
16. FDA (openFDA) 510(k) records search: records whose names include “da Vinci” (from K002489 in 2001 to K251213 in 2025; includes an unrelated device with the same name; checked 25 October 2026) — api.fda.gov/device/510k.json?search=device_name:%22da+vinci%22&limit=100&sort=decision_dat
17. Intuitive Surgical Form 10-K for the year ended 31 December 2025 (filed 3 February 2026; the company’s own description; checked 25 October 2026) — sec.gov/Archives/edgar/data/1035267/000103526726000010/isrg-20251231.htm
18. Medtronic Press release “Medtronic announces FDA clearance of Hugo robotic-assisted surgery system for urologic surgical procedures” (3 December 2025; the company’s announcement; checked 25 October 2026) — news.medtronic.com/2025-12-03-Medtronic-announces-FDA-clearance-of-Hugo-TM-robotic-assiste
19. Medicaroid / Kawasaki Heavy Industries / Sysmex Press release (11 August 2020; announcing Japanese regulatory approval of the hinotori Surgical Robot System; checked 25 October 2026) — global.kawasaki.com/news_200811-1e.pdf
20. Yang G.-Z. et al. “Medical robotics—Regulatory, ethical, and legal considerations for increasing levels of autonomy”, Science Robotics 2(4), eaam8638 (2017) — doi.org/10.1126/scirobotics.aam8638
21. Saeidi H. et al. “Autonomous robotic laparoscopic surgery for intestinal anastomosis”, Science Robotics 7(62), eabj2908 (2022) — doi.org/10.1126/scirobotics.abj2908
22. IEC IEC 80601-2-77:2019 “Medical electrical equipment – Part 2-77: Particular requirements for the basic safety and essential performance of robotically assisted surgical equipment” (IEC webstore page; full text is paid and was not read; checked 25 October 2026) — webstore.iec.ch/en/publication/29933
23. Miyasaka M. “Cable Driven Robots: Hysteretic Cable Stretch, Cable-Pulley Network Friction, Fatigue Life, and Kinematics of Two-Arm Multi Staged Flexible Manipulator” (University of Washington PhD thesis, 2017; based on the RAVEN II; checked 25 October 2026) — digital.lib.washington.edu/researchworks/handle/1773/40641

13. Claim-to-source audit

Last updated: 25 October 2026. Sources are limited to primary material (FDA pages and 510(k) / De Novo records, Intuitive Surgical’s Form 10-K, announcements by Medtronic and Medicaroid, Science Robotics papers, and an IEC standard page). Company figures and announcements are labeled as company claims and kept separate from regulator documents. The worked examples (the scaling ratio, the share of the installed base, the sum of study participants) are our own explanatory calculations and not the specifications of any device. This article is a general technical explainer and not medical advice. The scope of hinotori’s approval, the cable materials of each model and the numerical instrument-use limits could not be confirmed and are therefore not given. All figures are explanatory concept diagrams or drawings with our own calculations, except Figs. 1 and 6, which are ImageGen-generated conceptual illustrations. They do not depict real facilities, products or appearances.