top of page

Hot Mouse

MicroMouse Competition Team

Our goal is to build a MicroMouse robot capable of solving most competition mazes in under 3s, a world-record time, and then competing in the All Japan competition, the de-facto world championship. To achieve such a time, we are violating all current conventions by using brushless motors, custom ESCs, and time-of-flight sensors.

Status

Research Phase
We have finished researching components and planning which major components to use (sensors, motors, ICs).

Electrical
We have tested our electric circuits and components with a custom test PCB. Currently the Pico 2, custom ESC (voltage converter, hall effect sensors, motor driver, MOSFETS), motor, and IMU are working. The time-of-flight circuit still has some issues (chip works, but output is meaningless).

Mechanical
In terms of mechanical design, we have made a first prototype impeller (fan) and have a design for our custom gear train and motor mounts. We have already bought the tools necessary to machine the gears.

Next Steps
The next step is to fix the time-of-flight circuit, design the robot’s PCB layout, manufacturer the gear train and motor mounts, design a better impeller, and design the vacuum seal. Then assemble the robot, debug all the parts, and program it.

Context

Our Goal
Under 3 s time on most competition mazes.

Background
The MicroMouse competition has been around for a very long time (since the 1970s). In 2025 a classic-size maze was solved in 3.122 s at the All Japan competition. Typically the time is longer, but that year the maze was shorter than usual.

Other competitors have already pushed brushed motors to their limits. With robots as light as ~30 g achieving several g’s of acceleration, there is little to be innovated upon by copying other teams’ designs and making marginal improvements to the electrical/mechanical/algorithmic designs.

How do we plan to achieve this?
We are taking a very unconventional approach by using brushless motors, a custom ESC, and time-of-flight sensors. This is significantly more difficult because small brushless motors generally do not have encoders, they require an ESC, and control is more complex. However, the benefits are immense: the motors we chose are half the mass of the DC motors used by top competitors and have a power ~60x greater.

What can you learn?
While the task is more challenging, it is also very rewarding. To minimise mass and accommodate large currents on a tiny PCB, we’re using many components that have only been released within the last ~3 years. Many of these components do not have libraries written for the Pico (or Arduino), so we need to write them ourselves. We learn about many new things, from new components and tricky PCB design to exploring flexible, slippery materials for a partial vacuum seal.

Timeline

The Hot Beginnings

Jun

2025

We successfully finished our 2025 MicroMouse robot’s hardware and all its custom low level libraries by the end of RSA’s 2025 MicroMouse competition. After this unusual achievement, given we had hand-soldered dozens of QFN components without the right equipment, we realised we should aim higher.

We looked into the All Japan competition, the de-facto world championship, and were surprised to find that all competitors used expensive medical-grade DC motors. After some research, we concluded that pursuing a robot design with brushless motors and a custom ESC would allow for 10-100x higher mechanical power output. And with that, the journey towards a >500W Hot Mouse was born. We kept the same team: Sandro Karhula and Timon Riem.

Research Finished

Aug

2025

We finished researching motors, custom ESCs, various battery chemistries, supercapacitors (LICs, EDLCs, hybrids), and sensors. We decided upon most big components and purchased our first set of brushless motors. These motors have an electrical power input on the order of ~60x higher than the winning team in the 2025 All Japan competition, while being only half the mass.

Our Team Expands to 3

Oct

2025

Since this year will bring many new electrical challenges, we recruited Alexander Nitters to our team.

Custom PCB Arrives

Mar

2026

Our first custom PCB, designed to test all our components, arrives. Two days of soldering and debugging later, our Pico boots. Last year getting the Pico to boot took months.

The motor.... SPINS, HOT!

May

2026

Our voltage converter, custom ESC, hall effect sensor board, and motor finally come together, and the motor shows its first signs of life. Only days later with our first fan prototype we manage to pull 7N of downforce using a 40cm2 mouse surface area and 50W of power in a static test. Assuming similar dynamic performance, this is already sufficient to accelerate at ~4-7Gs, and it’s significantly more downforce than other top teams’ robots.

Recruitment Begins

Sep

2026

With the new academic year, Alex is starting his degree in Switzerland and will have limited time, while Timon has too many other responsibilities to be active. I remain active.

I have started looking for two new members: one with experience in electrical engineering and another in mechanical engineering.

UKMARS Competition

Apr

2027

The earliest competition that our robot may be able to compete in would be the UKMARKS MicroMouse competition organised every year in England in April. This competition would be useful to validate our design before competing in the All Japan competition (either in November or February of any year).

The Hot Beginnings

Jun

2025

We successfully finished our 2025 MicroMouse robot’s hardware and all its custom low level libraries by the end of RSA’s 2025 MicroMouse competition. After this unusual achievement, given we had hand-soldered dozens of QFN components without the right equipment, we realised we should aim higher.

We looked into the All Japan competition, the de-facto world championship, and were surprised to find that all competitors used expensive medical-grade DC motors. After some research, we concluded that pursuing a robot design with brushless motors and a custom ESC would allow for 10-100x higher mechanical power output. And with that, the journey towards a >500W Hot Mouse was born. We kept the same team: Sandro Karhula and Timon Riem.

Research Finished

Aug

2025

We finished researching motors, custom ESCs, various battery chemistries, supercapacitors (LICs, EDLCs, hybrids), and sensors. We decided upon most big components and purchased our first set of brushless motors. These motors have an electrical power input on the order of ~60x higher than the winning team in the 2025 All Japan competition, while being only half the mass.

Our Team Expands to 3

Oct

2025

Since this year will bring many new electrical challenges, we recruited Alexander Nitters to our team.

Custom PCB Arrives

Mar

2026

Our first custom PCB, designed to test all our components, arrives. Two days of soldering and debugging later, our Pico boots. Last year getting the Pico to boot took months.

The motor.... SPINS, HOT!

May

2026

Our voltage converter, custom ESC, hall effect sensor board, and motor finally come together, and the motor shows its first signs of life. Only days later with our first fan prototype we manage to pull 7N of downforce using a 40cm2 mouse surface area and 50W of power in a static test. Assuming similar dynamic performance, this is already sufficient to accelerate at ~4-7Gs, and it’s significantly more downforce than other top teams’ robots.

Recruitment Begins

Sep

2026

With the new academic year, Alex is starting his degree in Switzerland and will have limited time, while Timon has too many other responsibilities to be active. I remain active.

I have started looking for two new members: one with experience in electrical engineering and another in mechanical engineering.

UKMARS Competition

Apr

2027

The earliest competition that our robot may be able to compete in would be the UKMARKS MicroMouse competition organised every year in England in April. This competition would be useful to validate our design before competing in the All Japan competition (either in November or February of any year).

Open Roles

Electrical Engineer

We’re looking for an electrical engineer with expertise in PCB design.

You will work closely with an electrical engineer (the project lead) familiar with the current test PCB design.

Your main responsibility will be designing the new final PCB based on the existing test PCB. In addition, you will work with an electrical engineer to debug/fix any issues that come up in the designs. For this, competence in debugging circuits, datasheet comprehension, and basic soldering will be useful.

Mechanical Engineer

We’re looking for a mechanical engineer with CAD and machining experience.

You will take ownership of all the mechanical parts of the MicroMouse robot. You will work closely with the previous mechanical engineer who will introduce you to the current designs and explain the manufacturing process for the gear train.

Your main responsibilities will include manufacturing the gear train, designing the impeller (suction fan) and vacuum seal, and designing the motor mounts.

Application Details

We’re recruiting an electrical engineer and a mechanical engineer. Please send an email containing your background, relevant experience, and what you would like to work on.

Want to join the team?

Contact

To contact us, please send us an email at sandro.karhula@gmail.com

Application

Find us at

Website

LinkedIn

LinkedIn

GitHub

GitHub

Instagram

Instagram

Youtube

YouTube

The Team

Department

Name

Role

bottom of page