Field Notes · Robotics, Simulation & Surgical Technology

Robotic Surgery Explained: From Virtual Planning to Autonomy

Surgical robotics, virtual simulation and remote surgery solve different problems. Understanding who controls the instruments, what the software contributes and where the evidence comes from makes their future much easier to understand.

Who acts?

Identify the person or system controlling the instruments.

What changes?

Separate planning, visualization, training and physical action.

What is proven?

Keep each claim connected to its study and setting.

01 · Start with the vocabulary

“Virtual surgery” needs a more specific explanation.

A surgeon rehearsing in a simulator, a clinician reviewing a three-dimensional anatomical model and a surgeon controlling instruments from another location are doing different things. Describing all three as virtual surgery hides the distinction that matters most: what is happening to the patient, if there is a patient involved at all.

The following terms provide a more useful way to read an announcement or understand a demonstration. The mechanisms and research examples appear in the sections below.

Five terms, five different meanings
TermWhat it describesThe question to ask
Robot-assisted surgeryA surgeon uses a computerized system to control surgical instruments.Which task does the system help the surgeon perform?
Virtual simulationPractice or rehearsal in a computer-generated environment.Does performance transfer to the intended real task?
Virtual planning and ARDigital anatomical models or information added to a real-world view.Is this for visualization, planning or instrument navigation?
TelesurgeryA surgeon controls physical instruments from a distant location.How are the connection and local care arrangements supported?
Surgical autonomyA system executes defined actions without continuous manual control.Which actions, under what conditions and with what oversight?

These capabilities can coexist in one project. Combining them does not make them interchangeable. That distinction is useful across the wider science and technology landscape, where one dramatic label can conceal several separate engineering achievements.

02 · The person controlling the instruments

In conventional robot-assisted surgery, the surgeon remains in control.

The FDA’s explanation of computer-assisted surgical systems describes a common arrangement: a surgeon at a console views the surgical field and controls instruments mounted on mechanical arms. Software connects the controls with instrument movement. These surgeon-controlled systems do not independently decide how to perform the operation.

Potential advantages include helping with minimally invasive procedures and complex work in confined spaces. The relevant capability belongs to a particular device and use, rather than to the word “robotic” by itself.

To understand a demonstration, trace the decision backward. Did the surgeon command that movement? Did software execute a selected task? Did an algorithm choose the next action? The machinery can look similar while the division of responsibility changes substantially.

That is why a video of a robotic arm moving smoothly cannot, on its own, tell a viewer how autonomous the system is. Watching the instrument is only half the explanation; understanding the source of its instructions completes it.

03 · Three milestones worth knowing

Distance, training and autonomy developed along different paths.

  1. 2001

    An operation across the Atlantic

    On September 7, surgeons in New York operated on a patient in Strasbourg using the ZEUS robotic system. IRCAD’s account of Operation Lindbergh describes the landmark. The surgeon was distant; the operation and patient were physically present in a hospital.

  2. 2002

    Virtual practice put to a real test

    Seymour and colleagues tested whether simulator training changed residents’ operating-room performance. Their randomized study examined the practical transfer of learning.

  3. 2025

    Longer surgical steps performed autonomously

    The SRT-H team demonstrated autonomous surgical steps using gallbladders outside a living body. The experiment advanced task execution research.

The training and autonomy studies are linked below. Taken together, these milestones show why the field cannot be described as one steady progression from a human surgeon to a replacement machine. Different projects are changing different parts of surgical work.

04 · Learning before the operation

A simulator earns its value when learning transfers beyond the simulator.

In the 2002 randomized study of virtual reality training, 16 surgical residents were assigned to simulator training or a control group. The investigators then evaluated performance during laparoscopic gallbladder surgery, with observers blinded to training status. The trained group made fewer assessed errors.

The study involved one training approach and one operation. It provides evidence of skill transfer in that setting, rather than a performance guarantee for every simulator or curriculum.

For anyone evaluating a newer platform, I would start with three questions. What skill is being practiced? How does the learner receive useful feedback? What evidence connects the training with the intended work?

Visual realism can make a demonstration compelling, but resemblance and learning are different measures. A simulator should be assessed against its educational purpose. A convincing digital operating room is an interesting achievement; a demonstrated improvement in an important skill answers a different question.

05 · Seeing the anatomy differently

Virtual planning and augmented reality can change the view.

Virtual reality presents a simulated environment. Augmented reality adds digital information to a view of the real world. The FDA’s AR and VR overview describes medical applications that include planning and surgical visualization. A three-dimensional display can support understanding of anatomy without controlling an instrument.

A concrete example appears in the 2022 FDA clearance documentation for IntraOpVSP. Its stated uses include viewing and orienting selected three-dimensional anatomical models for planning and visualization. The same document explicitly excludes diagnosis, instrument guidance and surgical navigation from that intended use.

That boundary is useful well beyond one product. A model displayed near a patient and a navigation system guiding an instrument may look similar in a short video. They are different capabilities with different requirements. Read the intended use before deciding what the image demonstrates.

The source of the image matters too. Creating a model, presenting it to a user and acting on it are separate stages. Our guide to bio-optics in medical imaging explores the earlier question of how imaging methods obtain biological information.

A more informative display can be useful on its own terms. It does not have to become an autonomous surgeon to represent a meaningful advance.

06 · Moving expertise across distance

Remote surgery changes the surgeon’s location, not the patient’s need for care.

Operation Lindbergh established an important technical possibility: a surgeon could control instruments across a transatlantic connection. IRCAD’s account emphasizes overcoming transmission delay. The achievement depended on coordinated surgical, robotic and communications systems.

It helps to distinguish telesurgery from remote consultation and telementoring. A consultation exchanges information. A mentor can advise another clinician. In telesurgery, the distant operator controls physical instruments. Those differences change what must happen if the connection is interrupted.

For a proposed service, I would want the practical explanation to cover delay and its variation, communication loss, local clinical support and the transition to another workable arrangement. “Fast internet” leaves most of that explanation unfinished.

The relevant benefit is a possible change in access to expertise. Whether a particular service improves access, cost or outcomes has to be demonstrated in its actual setting. A successful connection cannot answer all three questions.

It also does not turn surgery into an at-home video appointment. In the transatlantic milestone, the patient remained in a hospital. Digital connectivity changed where the operator worked, while the physical requirements of the operation remained.

07 · Understanding the research frontier

Autonomy needs a task, a boundary and a setting.

The 2025 SRT-H paper in Science Robotics describes a hierarchy with a higher-level planner and a lower-level controller. The planner generated task instructions and corrective instructions; the controller produced robot trajectories. The team evaluated defined surgical steps using eight ex vivo gallbladders, meaning tissue outside a living body.

The researchers reported successful autonomous completion across those experiments. The paper characterizes the result as step-level autonomy. That finding should stay attached to the experiment: it does not establish a success rate for patients or demonstrate every responsibility involved in a complete clinical operation.

What interests me is the move from repeating a fixed motion toward managing a longer task and recovering when execution goes off course. The next questions concern how broadly that capability holds and how a system recognizes that it has reached the limits of its competence.

The task boundary

Specify what the system completed, what preparation was provided and which actions remained outside the experiment.

The evidence boundary

Identify whether evaluation used simulation, removed tissue, living animals or patients. Those settings answer different questions.

This is an important distinction in the broader advanced robotics industry. Autonomy is more informative when described as a specific capability than as a label attached to the entire machine.

08 · What better means in a clinical study

Outcome claims depend on the procedure, comparison and measurement.

Two trials illustrate why a general claim that robotic surgery is always better is unhelpful. They examined different operations and different outcomes. They are examples of how to read evidence, rather than a complete review of surgical robotics.

ROLARR · Published 2017

The ROLARR randomized trial compared robotic and conventional laparoscopic rectal cancer surgery in 471 participants. It did not find a statistically significant reduction in conversion to open surgery, its primary outcome. Participating surgeons had varying robotic experience.

iROC · Published 2022

The iROC randomized trial compared robotic bladder removal with intracorporeal reconstruction against open surgery. It found a statistically significant increase in days alive and out of hospital within 90 days. The authors stated that the clinical importance of the difference remained uncertain.

These findings are not competing verdicts on the same question. One comparison was against conventional laparoscopy; the other was against open surgery. The operations and primary outcomes differed as well.

My approach to a headline is to put those details back into the sentence. Better for which patients, undergoing which procedure, compared with what, and measured over how long? That fuller sentence is usually less dramatic and much more useful.

The same discipline belongs in discussions of medical devices and MedTech. Mechanical capability, workflow performance and patient outcomes should each be described with the evidence that actually addresses them.

09 · The operating room around the instrument

The people, interface and supporting systems remain part of the technology.

The FDA’s human factors and usability guidance focuses on intended users, uses and use environments, with the aim of reducing errors associated with device use. The interface and its setting belong in the evaluation alongside the hardware.

For a surgical system, useful review questions include who sees an alert, who understands the current operating mode and how a team communicates when responsibilities change. An impressive individual feature can still be difficult to use within a demanding sequence of work.

Connected equipment adds another layer. The FDA’s medical device cybersecurity overview explains that connectivity can introduce risks affecting device safety and effectiveness. It identifies manufacturers and healthcare organizations as participants in managing those risks.

From a systems perspective, I would want to understand support, maintenance, software changes and readiness for interruptions before treating a networked device as a dependable service. These practical details are where a technical capability meets the realities of healthcare delivery.

10 · Looking ahead without skipping the evidence

Useful progress may arrive one responsibility at a time.

My expectation is that progress will often be modular: a better training task, a clearer view, a more reliable connection or a carefully bounded automated action. Each can address a real problem without requiring a machine to take over the entire operation.

The research examples suggest several questions worth following. Can a training tool demonstrate transfer to clinical work? Can a planning display deliver information that helps its intended user? Can remote operation remain dependable in the proposed care setting? Can an autonomous task be evaluated across a wider range of relevant conditions?

Those are questions about usefulness, not just technical spectacle. A research demonstration opens a possibility; subsequent evaluation determines where that possibility becomes a dependable capability.

The field is fascinating because it brings perception, movement, software and human judgment into unusually close contact. The future is easier to appreciate when those contributions remain visible. There is plenty to be interested in without asking every new robot to graduate from the laboratory directly into its own operating room.

11 · Common questions

Robotic and virtual surgery FAQ

Is robotic surgery the same as autonomous surgery?

No. Conventional robot-assisted surgery uses instruments controlled by a surgeon. Autonomy describes a system executing defined actions without continuous manual control. The task and level of oversight need to be specified.

What does virtual surgery mean?

The term can refer to different activities, including simulation, digital planning or remote operation. A useful explanation states which activity is involved and whether physical instruments are acting on a patient.

How is augmented reality different from virtual reality?

Virtual reality presents a simulated environment. Augmented reality adds digital information to a real-world view. In surgical applications, the intended use may concern visualization or planning rather than controlling instruments.

Has surgery been performed across the Atlantic?

Yes. In Operation Lindbergh on September 7, 2001, surgeons in New York used a robotic system to operate on a patient in Strasbourg. The patient remained in a hospital.

Can virtual simulation help surgical training?

Evidence from specific studies supports particular training uses. A small randomized trial published in 2002 found fewer assessed operating-room errors after simulator training. That finding does not establish equal effectiveness for every training product.

Does remote surgery mean having an operation at home?

No. Remote describes the operator’s location relative to the patient. It does not remove the physical procedure or the need for appropriate local clinical care.

Have autonomous systems been proven for every part of surgery?

No. The SRT-H study discussed here demonstrated defined surgical steps using ex vivo tissue. That experiment does not establish autonomous performance across complete clinical operations in patients.

Does a surgical robot always improve patient outcomes?

No universal outcome follows from the robotic label. Assessment depends on the procedure, patient population, comparison, experience and outcome measured. A result for one clinical task should not be generalized to all surgery.

About the author

Robert Urban, Ph.D., is the founder of Paper Boat Media, an author and a scientist trained in Earth and Environmental Sciences. His work includes helping organizations explain complex science and technology. This article examines surgical technology and its evidence, with research and institutional sources linked throughout.

Scroll to Top