Camera Chase Vehicle

All<br>In-Progress<br>Photos<br>Panoramas<br>Vid<br>Blog<br>Social<br>People

Project Started: 01/2026

Building a Stabilized "Chase Camera" Rover

-->

You stumble on a weird robot chassis in an industrial warehouse. Its adorable, enormous and will likely be scrapped.

Time to adopt this curious contraption and turn it into something cinematic.

I was wandering around an industrial auction warehouse and stumbled on what looked like an enormous RC car with some elaborate scissor jack oddly attached to the top. The moment I spotted it I knew exactly what it should become: A distant off-brand cousin of the Freelfy Tero.

The Tero can be seen all the way back in the early rocketjump days [Link] where they used the platform to get low to the ground moving shots in their skits.

The following details all of the intricacies of building a chase camera from the ground up, using a mix of off the shelf items, used hardware and a pile of printed parts to bind them all together. There was a lot of trial-and-error in this build, and those errors are documented in full detail. As a result this writeup is a bit long and media heavy.

The Plan

Quad-rotor drone shots taken low to the ground are difficult: GPS altitude is fairly rough on accuracy, and obstacle avoidance can get significantly more difficult versus just flying over the everything. Cinema rover drones are less common but do get around a number of these problems, especially if the subjects are not high off the ground. As a fan of karting, lightweight contraptions and photography this seems like a pretty good project: Tackle a mechanically stabilized video platform, pilot it remotely and capture some outdoor action.

To do this we need three major things:

There's also A LOT of intermediate hardware to glue everything together, but the actual goal is to get shots where the subject is moving but in view.

This can be used for anything from zooming down a frozen lake on a madmaxian contraption or observing suspension geometry up close on a test track.

The Platform

A chassis that can hold a modern DSLR does need to be fairly large to support both a stabilization mechanism, like a 3 axis gimbal, while not flipping over.

Here's a 1/5 scale rc vehicle with a large dude for scale, used as marketing for Losi 5 platforms. These things are quite literally 1/5th the scale of a full size vehicle. The size both has advantages and disadvantages, namely everything is big and heavy. Big and heavy is great for preventing the whole platform from flipping but it does result in everything else becoming large, from the control electronics (VESC) to the comically large 3D prints.

The Gimbal

We all have that friend who forwards dangerously good deals on hardware, either oddities on marketplace or curiosities on Craigslist. I was fortunate enough to be linked to a very good deal on a completely functional Movi M10.Thanks Bayley.<br>This model was Freefly system's first go at a full size camera gimbal, proceeded by the M5 and then the M15. Either way for ~124 USD, it was a bargain and got some use at the 2025 Ice Racing Outing.

This Gimbal was the first born son of the Gimbal whisperer, Shane Colton, and it's very much a passion project. The fact that it's still being manufactured and iterated on, nearly a decade later, is a testament to how well it works.

The M10 is shown here with some support details here

To make development quicker, I'd really like to be able to remove the gimbal-camera-transmitter assembly as one unit, versus having a few dangling cables for power and communications. To do this we do need an auxiliary power supply. There is a happy medium between the gimbal's operating voltage range, the receiver's operating voltage range, the camera's aux power supply voltage range and the wireless video transmitter's voltage range, and its a convenient ~14V, or rather, the DTAP battery range of 12-16.8Vv

While there are a lot of things to power, there's a bit of attention to detail to ensure the total mass is kept low, especially on the most actuated portion of the gimbal.

Radio and Telemetry

This cart needs to relay back video information over a long distance. Traditionally this can be a game of quality versus latency vs bandwidth. I'm also limited to what my video source is, which likely a DSLR with an HDMI output. Traditionally I've used the 900 MHz band for long range control, but for video, 5.8 GHz really does seem appropriate. I was able to find a matching pair of Amimon CONNEX wireless HD links intended for UAV's for

Upgrade-ability

While the goal for this project is to get something working quickly, allocating space and planning accordingly for integrating some companion computation for automated tracking would be excellent. For the pixhawk system, this means having enough space and mounting points for a single board computer for vision processing and or visual based obstacle avoidance. One of the best ways to have space going forward is to have an accurate...