TURN A WHEELCHAIR INTO ONE YOU DRIVE WITH YOUR HEAD

A power wheelchair that drives on head tilts and navigates learned routes by voice command.

by HTX Studio

RoboticsHealth

difficulty
●●●●●
time
months of integration work
license
license not specified
repo
repo
1
Jump to section

COMPAREE VERDICT

This is not a build guide. It is documentation of a one-off commissioned assistive device that HTX Studio built for a man paralyzed from the neck down. The wheelchair responds to head tilts: down for forward, up for reverse, turning to steer. A high-precision gyroscope on the base compensates for the chair's own rotation, which otherwise caused the system to spin in circles. Autonomous navigation uses lidar repurposed from HTX's self-driving office chair; the system learns routes by being pushed through them once. Voice commands then trigger navigation to saved waypoints. The video is a demonstration, not a tutorial. There is no repository, no parts list, no code, no STL files, no PCB design, and no instructions. Someone wanting to replicate this would be reverse-engineering from footage alone, which means sourcing a head-tracking sensor system, a gyroscope with sub-degree precision, a lidar unit with SLAM capability, integrating all three into a power wheelchair's existing control system, and writing the sensor fusion and path-planning code from scratch. This is a months-long robotics integration project requiring embedded systems experience, ROS or equivalent middleware, and access to a power wheelchair you are willing to modify. It is not a weekend build. It is also not certified for medical use. If you need an assistive device, speak to an occupational therapist and a medical equipment supplier first.

GOOD TO KNOW

  • —This is a commissioned modification by HTX Studio, not a published open-source project.
  • —No repository, no build guide, no schematics, no code.
  • —Requires a store-bought power wheelchair as the base platform.
  • —Integrates head-tracking sensor hardware, precise gyroscope, lidar unit, voice recognition, and custom control firmware.
  • —Not a certified medical device.
  • —The video shows the result, not the build process.

Parts to buy

7 items

From our check of the build. Exact quantities and part numbers are in the creator’s BOM.

  • Store-bought power wheelchairFind
  • Head-tracking sensor hardwareFind
  • High-precision gyroscopeFind
  • Lidar unit with SLAM capabilityFind
  • Embedded compute platformFind
  • Microphone for voice controlFind
  • Wiring and power distributionFind

Can I build this?

Printnothing required
Buystore-bought power wheelchair, head-tracking sensor hardware, high-precision gyroscope, lidar unit with SLAM capability, embedded compute platform, microphone for voice control, wiring and power distribution
Toolsrobotics middleware (ROS or equivalent), embedded development environment, soldering and electronics assembly tools, wheelchair modification tools, test environment with space to debug autonomous navigation safely
Skillsembedded systems programming, sensor fusion, SLAM and path planning, ROS or middleware integration, power wheelchair electronics, accessibility device safety protocols
Timemonths of integration, testing, and tuning for a single user
Cost$$$, dominated by the power wheelchair base and the precision gyroscope
SafetyModifying a power wheelchair for autonomous control creates serious collision and injury risk during development and use. This is not a toy and must be tested in a controlled space. Autonomous navigation near stairs, obstacles, or other people requires fail-safes. This project bypasses certified medical device standards. If building for yourself or another person, consult medical and legal professionals first.

Build at your own risk. Projects involve tools, electronics and sometimes mains voltage — follow the creator’s safety notes.

Videos

Head-controlled wheelchair demonstration7:31

Full demonstration and explanation by HTX Studio. Not a build tutorial.

More builds like this

All projects

Gallery

Start here

Navigation into the creator’s own docs — we don’t rewrite the guide, we route you to the source.

  1. 1.Understand the scope (Watch the full video to see what the system does. This is a demonstration, not a tutorial.)
  2. 2.Assess your requirements(Is this for yourself, for someone else, or for learning? If it is for assistive use, speak to an occupational therapist and a medical equipment supplier before modifying a wheelchair.)
  3. 3.Source the base platform(You need a power wheelchair with accessible motor control signals. Most modern wheelchairs use proprietary control systems that are difficult to interface with.)
  4. 4.Research sensor options(Head-tracking sensors exist for gaming and accessibility (e.g. TrackIR, Tobii head trackers). A gyroscope with sub-degree precision is harder to source affordably. The lidar could be a robotics lidar such as RPLidar or Slamtec, but you must verify SLAM capability.)
  5. 5.Plan the control architecture(You are building a sensor fusion system that reads head angle, subtracts wheelchair rotation, and outputs motor commands. ROS is the standard middleware for this, but the learning curve is steep.)

KNOWN ISSUES

  • Most power wheelchairs have proprietary control systems that do not expose motor control signals. You may have to reverse-engineer the joystick interface or replace the entire control board, which voids warranties and may violate medical device regulations.
  • A gyroscope that drifts even slightly will cause the chair to veer or spin over time. High-precision IMUs cost hundreds of dollars, and you still need sensor fusion code to combine them with the head tracker.
  • Autonomous navigation in a home is harder than it looks. The lidar must build a map, localize within it, plan collision-free paths, and re-plan when furniture moves. Existing SLAM libraries exist, but integrating them with real-time motor control is not trivial.
  • Teaching the chair a route by pushing it once only works if the lidar's localization is accurate to within a few centimeters. In practice, this requires careful tuning and repeated test runs.
  • Voice recognition in a noisy home is unreliable without wake-word detection and noise cancellation. Off-the-shelf solutions exist, but they add latency and require cloud connectivity or local speech models.
  • This is a one-user system. Head-tracking thresholds, turning sensitivity, and route waypoints are tuned for one person. A second user would need recalibration.

Where is the build guide?

There is none. This was a commissioned project by HTX Studio, not an open-source release. The video shows the result, not the build process.

What head-tracking sensor did they use?

The video mentions a sensor on the user's ear, but does not name the model. You would have to research accessibility head trackers or contact HTX Studio.

Can I use a cheaper gyroscope?

Not if you want the same result. The video specifically mentions a high-precision gyroscope to subtract the chair's rotation. Consumer IMUs drift too much for this application.

Is this legal to build?

That depends on your jurisdiction. Modifying a power wheelchair may void its certification as a medical device. If you are building it for yourself, the legal risk is lower than if you are building it for someone else. Consult a lawyer if this matters to you.

How much would this cost to replicate?

Thousands of dollars. The power wheelchair alone is $1,000-5,000 depending on the model. A precision IMU is $200-500. A robotics lidar is $100-400. Embedded compute, sensors, and wiring add another few hundred. Then there are months of your time.

Could I sell this?

Not without certifying it as a medical device, which is a multi-year regulatory process. Selling an uncertified assistive device exposes you to serious liability.

Community builds

No community builds yet — be the first, we feature the best ones.

Discussion1

FROM THE COMPAREE TEAM

HTX repurposed the lidar from their self-driving office chair and added a high-precision gyroscope to stop the spinning problem. If you were building this, what sensor would you start with: the head tracker, the gyro, or the lidar?

CompareeTEAM20d ago

Practical notes from our verification: this is the only public documentation of the project, there is no repository or build guide, and HTX Studio has not released the parts list or code. The video is a demonstration, not a tutorial. The single biggest challenge if you wanted to replicate this is interfacing with the power wheelchair's existing control system, which is proprietary and not designed to accept external inputs. The autonomous navigation is impressive, but it only works for learned routes in a known environment, not for navigating arbitrary destinations. The user in the video says he can go wherever he wants, but that means saved waypoints in his home, not outdoor navigation or new buildings.

HTX Studio

HTX Studio is a design and engineering channel that builds autonomous systems and assistive devices. The user in the video, a former developer of airborne hardware, was paralyzed from the neck down in an ATV accident in August 2024 and asked HTX for a head-controlled wheelchair. HTX had previously built a self-driving office chair, and they repurposed the lidar from that project for this one.

Star the project on GitHub

DISCLAIMER

  • Comparee is not the author of the projects featured here. All rights to each project belong to its creator — every page links to the original source, and we never host creators’ files.
  • Information is provided without warranty and may become outdated as projects evolve. Prices are indicative bands only — always check the creator’s parts list for current costs.
  • Building and operating any project is at your own responsibility. Protective equipment, safe workshop practice and compliance with local regulations are the builder’s responsibility.