Swimming

Swimming
Load video YouTube YouTube might collect personal data. Privacy Policy ContinueDismiss
Information
Type	Seated, Standing, Room-scale
Required Devices	Motion-tracked controllers
Description	Hold the triggers on the controllers then stroke or paddle to move.
Pros	Arm strokes give the user a physical action that matches the movement; in a seated study it reduced VR sickness about as much as teleportation while being rated more enjoyable than teleportation
Cons	Mostly used for water and underwater environments; tires the arms and shoulders; slower than joystick or teleport movement in some studies and rated harder to use
Examples	Ocean Rift, Freediver: Triton Down

Swimming is a Locomotion Method for virtual reality (VR) in which the user moves through a virtual environment by making swimming strokes with the arms, usually while holding tracked motion controllers. The software measures the motion of the hands during each stroke and moves the user's viewpoint in response, so faster strokes produce faster travel and the direction of the stroke sets the direction of travel. In the controller-based versions described in research, the user holds the triggers during a stroke, so ordinary hand movements do not move the player.

The technique is used mainly in underwater settings, where swimming is the expected way to move, and it can move the user's viewpoint in all directions rather than only across a floor. Research has also tested swimming-style arm motions as a seated form of artificial locomotion meant to reduce VR sickness outside water-themed content.

Reviewed 27 September 2026. Checked every claim against the cited Huang et al. 2019 and Shimoyamano and Arakawa 2026 papers, the Seno 2013 and Uchida 2021 abstracts, the Fauville 2021 article, the AquaCAVE project page, the Ocean Rift Mod DB report and Steam page, and the Road to VR and UploadVR Freediver articles. About review dates.

How it works

In a typical controller-based implementation, the user holds a controller in each hand, pulls the triggers at the start of a stroke and releases them at the end. The movement of the controllers during the stroke is mapped to movement of the user's avatar, and the start and end points of each stroke steer the user. In the underwater game Freediver: Triton Down, the player pulls the triggers and swings both arms together to push in the desired direction, while a stroke with only one arm rotates the body to turn.

Some systems also try to reproduce the feel of moving through water. The research system Swit Loco, for example, applies inertial processing to its movement and rotation to simulate underwater movement. Swimming locomotion is not limited to controllers: in an experiment by Seno, Funatsu and Palmisano, a Kinect sensor tracked a standing observer's breaststroke arm movements and used them to generate the optic flow shown on the display.

Research

Vection

Seno, Funatsu and Palmisano named the effect of active breaststroke movements on illusory self-motion "virtual swimming". In their 2013 study, both congruent strokes (optic flow matching the observer's head motion) and incongruent strokes increased vection compared with passively watching the same optic flow, and congruent strokes increased it more than incongruent ones. The authors concluded that even unusual body movements can act as a self-motion signal.

VR sickness

Kenta Uchida and Tokiichiro Takahashi reported at the International Workshop on Advanced Imaging Technology (IWAIT) 2021 that a sensory interface they built to mimic the "swimming motion" gave good results in reducing VR sickness. Their paper compared users' sickness from the swimming motion with their general motion sickness susceptibility, measured with the Motion Sickness Susceptibility Questionnaire (MSSQ).

Rei Shimoyamano and Toshiya Arakawa proposed Swit Loco, a seated locomotion method based on swimming, in a paper published in January 2026. When asked to move a caster-equipped chair using only their upper body, several participants swung their arms in a breaststroke-like motion, and this observation led to the design. In Swit Loco, sweeping both hands horizontally around the body moves the user forward, swinging both hands forward moves the user backward, and throwing both hands backward in the same rotation turns the user; locomotion is active only while both index triggers are held. The authors built it in VRChat and tested it with a Meta Quest 2 connected to a PC. Fifteen male participants aged 21 to 47, who had no prior VR experience or only one previous exposure, completed a shooting course designed to take about two minutes with Swit Loco, VRChat's joystick movement and VRChat's "holoport" teleport. Scores on the Virtual Reality Sickness Questionnaire (VRSQ) were significantly lower for Swit Loco and for teleportation than for the joystick, and the authors judged Swit Loco similar to teleportation in reducing sickness. It scored higher than teleportation on most quality-of-experience items, including enjoyment, but lower on familiarity. The authors noted that the sample was small, all male and limited to VRChat.

Comparisons with other methods

A 2019 study from UC Berkeley and the Lawrence Hall of Science compared four locomotion methods in an underwater task where participants moved to and touched sea animals. The methods were swimming, "Look & Follow" (moving in the direction the head faces, with speed set by pulling a trigger), teleportation, and an "Assisted Swimming" mode in which participants made swimming strokes but were actually moved by the Look & Follow method without being told. The researchers kept data from 40 adults and 40 minors, with 10 of each per method. Swimming was the slowest method for both groups (average task times of 276.00 seconds for adults and 286.60 seconds for minors), and both groups rated it the most challenging. Adults rated it their least preferred method (4.90 out of 10), and minors gave it their lowest preference score (6.40). Minors completed the task fastest with Assisted Swimming and rated it their favorite method (8.20). The authors concluded that physical body movement that exclusively mirrors real-world movement was the least preferred approach for both adults and minors.

Problems

Swimming locomotion demands continuous arm movement. In the Swit Loco study, participants' written comments included "My arm gets tired" and "My shoulders ache", and the authors listed arm fatigue as a weakness of the method. Some participants in that study also found the turning motion and its direction confusing, reported moving or turning at speeds they had not intended, or moved in the opposite direction to the one they meant, and some still felt motion sickness while turning. In the Berkeley study, swimming took the longest of the four methods to complete the task and was rated the most challenging. Because it occupies the user's hands, swimming can also conflict with other hand-based interactions in the same application.

Solutions

Requiring the user to hold both triggers before a stroke counts keeps the hands free for other tasks the rest of the time; Swit Loco activates locomotion only while both index triggers are held so that the method does not interfere with the shooting tasks in its test scenario. Freediver: Triton Down lets players adjust how much speed and momentum each stroke gives; Road to VR's Ben Lang found the default to be a good balance between realism and not feeling like a workout. Hybrid methods separate the arm motion from the actual steering: the Berkeley study's Assisted Swimming mode kept the swimming gestures to add immersion but moved the user with head-directed movement, and minors rated it their favorite of the methods tested. The 2016 HTC Vive prototype of Ocean Rift added a small propeller that could be extended from one of the controllers and toggled on and off for additional directional control. The Swit Loco authors suggested that the method's movement performance may need to be personalized for each user.

Use in games

Ocean Rift, an underwater VR experience by the Welsh developer Llyr ap Cenydd, launched first on the Samsung Gear VR. A January 2016 Mod DB report described a video in which ap Cenydd demonstrated an HTC Vive prototype where swimming motions made with the motion controllers moved the player's first-person view in all directions. The Steam version, developed by ap Cenydd and published by Picselica Ltd, was released on 13 September 2017.

Freediver: Triton Down, developed by Archiact, is an underwater adventure game built around arm-stroke swimming, with a wrist-mounted oxygen gauge that keeps the player under pressure while exploring and solving puzzles. In UploadVR's review of the PC VR version, Jamie Feltham wrote that "swimming feels almost as natural as zero gravity traversal with a few of its own twists" and described the mechanic as "intuitive and, crucially, immersive", comparing it with the zero-gravity movement of Lone Echo. An "Extended Cut" with new levels was released for PlayStation VR and Oculus Quest on 25 February 2020; the PlayStation VR version supports only PlayStation Move controllers.

Swimming in real water

A separate line of work puts VR users in actual water. AquaCAVE, developed at Jun Rekimoto's laboratory at the University of Tokyo, is a swimming pool with rear-projection acrylic walls that surround the swimmer, giving a CAVE-like stereoscopic projection environment. The swimmer wears goggles with LCD shutter glasses while infrared cameras track the head, and the system can show scenes such as coral reefs or outer space, or training aids such as record lines. It was presented at UIST 2016 and at ICAT-EGVE 2016.

Head-mounted displays have also been used in pools. In a 2021 study in Scientific Reports, Geraldine Fauville, Jeremy Bailenson and colleagues compared an underwater drift-dive simulation viewed on land in a Samsung Gear VR with the same content viewed in a pool through a waterproof DIVR headset made by Ballast Technologies. The DIVR headset puts water between the lenses and the display to reduce the headset's buoyancy and keep the image clear underwater; in-pool participants floated face down with a snorkel and a flotation belt, held in place by an elastic tether to an anchor on the pool floor. Vection was significantly stronger in the water condition, with no difference in motion sickness or presence between conditions. The authors noted that participants' behavior ranged from floating motionless to actively swimming, although they did not analyze these behaviors.

For augmented reality during real swimming rather than VR locomotion, see Guangli Holoswim.