Freedom Locomotion System
More actions
| Freedom Locomotion System | |
|---|---|
| Information | |
| Type | Standing, Room-scale |
| Required Devices | Motion-tracked controllers |
| Description | Collection of various standing and room-scale locomotion solutions with CAOTS as its foundation. |
| Pros | Hands stay free for interaction while moving; travel direction is independent of head direction; physical walking in place is intended to reduce motion sickness compared with sliding locomotion; teleport-style fallbacks for sensitive users |
| Cons | Walking in place is physically tiring; users drift forward in the real play area over time; setting direction with the thumb adds an artificial step that new users forget; a learning period of several minutes |
| Creator | HUGE ROBOT |
| Examples | Freedom Locomotion VR (demo) |
The Freedom Locomotion System (FLS) is a Locomotion Method for virtual reality developed by George Kong of the Australian studio Huge Robot and released in 2017 as the free demo Freedom Locomotion VR. It combines several ways of moving through a virtual world into one package. Its foundation is CAOTS (Controller Assisted On The Spot), a walking-in-place technique in which the user walks, jogs or runs on the spot to move forward while a thumb on the controller's touchpad or thumbstick sets the direction of travel.[1][2][3]
Alongside CAOTS, the system offers two teleport-derived modes (Dash Step and Blink Step) for users who prefer less physical effort or are more prone to motion sickness, plus sub-systems for turning, climbing and handling collisions with virtual walls.[4] It needs no hardware beyond a VR headset with tracked motion controllers.[5] The demo launched on itch.io for the HTC Vive in January 2017 and on Steam with Oculus Rift and Oculus Touch support on 13 March 2017.[6][7]
History
Huge Robot described itself as a VR design and development consultancy; according to the studio's website it was a one-person operation run by George Kong in Perth, Western Australia. Kong, a designer and 3D modeller, wrote that he had been working on a book about "The Future of Virtual Reality" and paused it to learn programming in Unreal Engine 4, after which he devised CAOTS and then built the other components around it.[8][9]
Road to VR covered the system on 11 January 2017 with a video breakdown by Kong, who said the package was "as close to solving the issue of immersive VR locomotion as we can get within the current practical limitations of VR."[2] Kong announced the free Freedom Locomotion VR demo on itch.io in a blog post dated 17 January 2017. He called it "just a tech demo", said it worked with the HTC Vive, and said Rift compatibility would follow. He also wrote that he had received many requests to release the system as a plugin, but that making it ready for other developers' games would need refactoring, optimization and documentation.[6][1]
Early feedback led to changes. In a 26 January 2017 post Kong wrote that many users found walking movement not sensitive enough, because walking styles differ a lot once the horizontal part of the motion is removed, and that some users still became motion sick on slopes and when falling. His response was a per-user calibration system and a rebuilt field-of-view restriction that narrows more when the motion is expected to cause more vection.[10] The Steam release in March 2017 added full Rift and Touch compatibility and a variable-angle snap turn.[7] In that post Kong said he planned to add sliding locomotion and a proxy-based teleportation mode, and to turn the system into a plugin for Unreal Engine first and Unity later.[7] Sliding locomotion arrived as a fourth mode in version 1.3.1 on Steam in May 2017.[11]
Version history
| Version | Date | Main changes |
|---|---|---|
| 1.0 | January 2017 (itch.io) | First public demo for the HTC Vive, with CAOTS, Dash Step and Blink Step[12][6] |
| 1.1.0 | 1 February 2017 | Detailed calibration of CAOTS movement ranges; sharper, adjustable comfort FOV restriction; "slide to a stop" deceleration; climbing feedback (yellow and green hand outlines, audio cues); redesigned boundary violation handling[12] |
| 1.1.1 | 2 February 2017 | Walking algorithm tuned for slow, sneaking movement[12] |
| 1.2.0 | 13 March 2017 | Oculus Rift and Touch compatibility; Steam release; arbitrary-angle snap turning; revised Dash and Blink Step speed control; optional fade to black for Blink Step[12][7] |
| 1.3.1 | 4 May 2017 (Steam) | Sliding locomotion added as a fourth mode; optional hybrid CAOTS and arm-swinging input; moving platforms; experimental rectangular comfort FOV; improved climbing and gun recoil[11] |
How it works
CAOTS reads the up-and-down bob of the headset together with controller input. The user places a thumb on the touchpad (Vive) or pushes the thumbstick (Touch) in the desired direction and then walks, jogs or runs in place; moving faster in place moves the user faster in the virtual world.[1][12][13] Because direction comes from the controller rather than the headset, the user can look around while moving, and because the hands do not have to swing, they stay free to hold a gun or sword.[1][2] Huge Robot's project page explains the use of the controller by arguing that walking-in-place systems cannot track a user's intent perfectly from headset motion alone, or even from foot motion as with omnidirectional treadmills.[4]
From version 1.2.0 the Vive version defaults to a "Simple Touch" mode, in which the controller's pointing direction sets the heading and the touchpad only chooses forward or backward. Movement on the Vive starts with a click and continues while the thumb stays on the touchpad, which Kong said reduced strain on thumbs and on the touchpad itself.[12][7] A first-run calibration asks for the user's height and play-area size, then has the user walk, jog and run so the system can set its detection thresholds.[14]
The system also moves the user up and down with slopes and stairs walked in physical space, instead of letting the view clip through geometry.[2] A field-of-view restriction comfort option narrows the view during artificial movement; its comfort level is chosen during calibration, and comfort boundaries stay visible through the black borders.[12]
Components
Huge Robot describes the system as three locomotion methods plus three sub-systems.[4] Sliding (smooth) locomotion was added as a fourth method in version 1.3.1.[11]
- CAOTS - Controller Assisted On The Spot - running-in-spot method that allows the user to walk, run and sprint in VR by performing the same movement in a spot in real life. Players use the touchpad on the controller to change the direction of movement in VR. Additionally, players can also swing their arms higher to move faster. CAOTS differentiates itself by having accurate representation of head movement in VR. It also allows the user to still use their hands to interact with objects while moving. Version 1.3.1 added an optional hybrid mode in which arm swinging, with one or both hands, can start and speed up motion.[11]
- Dash Step - Press touchpad to move a short distance forward. Hold down the touchpad to move forward at a steady pace. Press the touchpad faster to move faster. That was the control scheme of the first release. Version 1.1.0 changed it to a single press with the thumb kept on the touchpad, and version 1.2.0 replaced faster clicking with arm position: the angle of the elbow and hand sets the step speed and the distance of the hand from the body sets the step length.[12][7] Huge Robot describes the mode as short dashing teleports that mimic taking a step.[4] UploadVR compared it to the dash movement in Raw Data.[1]
- Blink Step - Press touchpad to teleport a short distance forward. It is for users who experience VR sickness from Dash Step. It works like Dash Step but replaces the visible motion between steps with a pause of equivalent length, and version 1.2.0 added an optional fade to black.[4][12]
- Grab Turning - Allows the user to turn in VR without turning in real life by grabbing and turning the VR environment with the controller. This is very useful for front-facing VR systems such as Oculus Rift and PlayStation VR. The project page says it was first conceived as a way to realign the virtual space with the physical play area.[4] Version 1.2.0 added a separate snap turn in which the user points the controller in the new direction and presses a button; unlike fixed-angle snap turns, the rotation angle is variable.[7]
- Anti-Boundary Violation - When you try to walk through virtual walls, screen will darken and Chaperone boundaries appear. When you walk over virtual ledge, you will fall in VR. Obstacles the user is intruding into are shown in red while the rest of the view blacks out.[4] From version 1.1.0 the system mainly acts on movement in room space, so running into a wall with CAOTS no longer teleports the user back.[12]
- Free Climbing - Procedural climbing system that doubles as a crawling system. What can be grabbed is determined from the kind of handhold a 3D mesh provides, so nearly anything that looks climbable can be climbed.[4][2] The user grips an edge by closing the virtual thumb and forefinger around it.[10]
Freedom Locomotion VR demo
The demo includes a tutorial, three levels and a hub level called "The Nexus" that links them and holds collected items.[4] Its terrain ranges from open abstract spaces in the sky to a 500-meter rocky mountain trail, and it includes test objects such as swords and guns for trying the system while handling objects.[12] The developer estimated the learning curve at 5 to 20 minutes.[4] The Steam listing names Huge Robot as developer and publisher, gives 13 March 2017 as the release date, and lists the demo as free, VR-only and SteamVR-based, with a minimum of an Intel i5-4590, 4 GB of RAM and an Nvidia GTX 970.[3] The itch.io page lists version 1.2.0 (13 March 2017) as its latest download.[12]
Reception and limitations
Writing for UploadVR, David Jagneaux said the controller-plus-head-bob approach set CAOTS apart from other running-in-place systems he had tried, and that among the locomotion systems the site had covered, such as PocketStrafe, the 3DRudder and ArmSwinger, Freedom Locomotion "seems to present the most promise thus far."[1] KitGuru reported very little motion sickness in its testing and called it "the most effective locomotion system I have tried to date," but found that faster movements tracked better than slow ones, that a "boxer's shuffle" tracked better than walking in place, and that side-to-side movement at high speed could still be nauseating.[13]
In a March 2017 review on The Ghost Howls, written using an Oculus Rift with Touch by a developer whose own company had built the ImmotionRoom walking-in-place system, the reviewer judged walking detection reliable and sensitive enough that only light movement was needed. The same review listed drawbacks: walking in place is tiring over long sessions; users drift forward in the real room and must reposition to stay inside the boundary; new users often walk in place but forget to push the direction control; and walking in one direction while steering sideways with the thumbstick still caused motion sickness.[14] The reviewer found Dash Step and Blink Step less tiring but "not natural at all".[14] As of 27 September 2026, the Steam page showed a "Very Positive" rating, with 86% of 222 user reviews positive.[3]
Background: walking in place
Walking in place as a VR travel technique predates FLS by more than two decades. Slater, Usoh and Steed (1995) used a neural network to detect walking on the spot from head-mounted display tracking data and moved the user in the direction of gaze; their studies suggested that it increased subjective presence compared with hand-pointing navigation, provided that participants identified with their virtual body.[15] Usoh and colleagues (1999) replicated that study with real walking added as a third condition. Both kinds of walking produced higher subjective presence than push-button flying, and real walking scored higher than walking in place, although that difference was statistically significant only in some models. They also reported that follow-on studies improved walking in place substantially by detecting footfalls with a head-mounted accelerometer.[16] Templeman, Denbrook and Sibert's Gaiter (1999) sensed leg movement to set the direction, length and timing of virtual steps.[17] CAOTS differs from the Slater technique in taking direction from a hand controller rather than gaze, and from Gaiter in needing no leg sensors.[15][13][17][5]
See also
References
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 David Jagneaux (2017-01-17). "'Freedom Locomotion' Wants to Be Your Fully-Featured Movement System for VR Games". UploadVR. https://www.uploadvr.com/freedom-locomotion-vr-movement-system/. Retrieved 2026-09-27.
- ↑ 2.0 2.1 2.2 2.3 2.4 Ben Lang (2017-01-11). "'Freedom Locomotion System' is a Comprehensive Package for VR Movement". Road to VR. https://www.roadtovr.com/freedom-locomotion-system-comprehensive-package-vr-movement/. Retrieved 2026-09-27.
- ↑ 3.0 3.1 3.2 "Freedom Locomotion VR on Steam". Steam. Valve. https://store.steampowered.com/app/584170/Freedom_Locomotion_VR/. Retrieved 2026-09-27.
- ↑ 4.00 4.01 4.02 4.03 4.04 4.05 4.06 4.07 4.08 4.09 "Projects". Huge Robot. 2017. https://web.archive.org/web/20170906085411/https://hugerobotvr.com/projects/. Retrieved 2026-09-27.
- ↑ 5.0 5.1 "Huge Robot". Huge Robot. 2017. https://web.archive.org/web/20190717063043/https://hugerobotvr.com/home. Retrieved 2026-09-27.
- ↑ 6.0 6.1 6.2 George Kong (2017-01-17). "Demo is launched". Huge Robot. https://web.archive.org/web/20190717071733/https://hugerobotvr.com/blog/2017/1/17/demo-is-launched. Retrieved 2026-09-27.
- ↑ 7.0 7.1 7.2 7.3 7.4 7.5 7.6 George Kong (2017-03-13). "Steam launch of Freedom Locomotion VR with Oculus Rift + Touch compatibility". Huge Robot. https://web.archive.org/web/20190717072113/https://hugerobotvr.com/blog/2017/3/10/steam-launch-of-freedom-locomotion-vr-with-oculus-rift-touch-compatibility. Retrieved 2026-09-27.
- ↑ "About". Huge Robot. 2017. https://web.archive.org/web/20190717065907/https://hugerobotvr.com/about. Retrieved 2026-09-27.
- ↑ George Kong (2017-01-04). "The first blog post!". Huge Robot. https://web.archive.org/web/20190717062846/https://hugerobotvr.com/blog/2017/1/4/the-first-blog-post. Retrieved 2026-09-27.
- ↑ 10.0 10.1 George Kong (2017-01-26). "Working hard on improving the Freedom Locomotion System". Huge Robot. https://web.archive.org/web/20190717070258/https://hugerobotvr.com/blog/2017/1/26/working-hard-on-improving-the-freedom-locomotion-system. Retrieved 2026-09-27.
- ↑ 11.0 11.1 11.2 11.3 "Freedom Locomotion VR updated to v1.3.1". Steam Community. Valve. 2017-05-04. https://steamcommunity.com/games/584170/announcements/detail/1304200199069334426. Retrieved 2026-09-27.
- ↑ 12.00 12.01 12.02 12.03 12.04 12.05 12.06 12.07 12.08 12.09 12.10 12.11 "Freedom Locomotion VR by HugeRobot". itch.io. https://hugerobot.itch.io/freedom-locomotion-vr. Retrieved 2026-09-27.
- ↑ 13.0 13.1 13.2 Jon Martindale (2017-01-18). "VR locomotion solved? The CAOTS system". KitGuru. https://www.kitguru.net/components/vr/jon-martindale/vr-locomotion-solved-the-caots-system/. Retrieved 2026-09-27.
- ↑ 14.0 14.1 14.2 Skarredghost (2017-03-28). "Freedom Locomotion System Review". The Ghost Howls. https://skarredghost.com/2017/03/28/freedom-locomotion-system-review/. Retrieved 2026-09-27.
- ↑ 15.0 15.1 Mel Slater, Martin Usoh, Anthony Steed (1995). "Taking steps: the influence of a walking technique on presence in virtual reality". ACM Transactions on Computer-Human Interaction, vol. 2, no. 3. https://doi.org/10.1145/210079.210084. Retrieved 2026-09-27.
- ↑ Martin Usoh, Kevin Arthur, Mary C. Whitton, Rui Bastos, Anthony Steed, Mel Slater, Frederick P. Brooks (1999). "Walking > walking-in-place > flying, in virtual environments". Proceedings of SIGGRAPH '99. https://doi.org/10.1145/311535.311589. Retrieved 2026-09-27.
- ↑ 17.0 17.1 James N. Templeman, Patricia S. Denbrook, Linda E. Sibert (1999). "Virtual Locomotion: Walking in Place through Virtual Environments". Presence: Teleoperators and Virtual Environments, vol. 8, no. 6. https://doi.org/10.1162/105474699566512. Retrieved 2026-09-27.