Veter Veter - robotics vehicle for researchers and makers




Main steps

We are constantly developing and improving our robot. As a result, the building instructions could quickly become outdated. However, our project wiki is kept up to date. That is why we decide to provide here high-level overview of required steps and refer to the corresponding wiki articles for more details.

If you decide to build the robot yourself, here are the main steps to accomplish:
  1. 3D print the robot's body. Currently we are using Ultimaker 3D printer. All the models (original Blender files and generated STL files) are freely available from our git repository.
  2. Buy electronic and mechanical parts. Our goal is to make electronics and mechanic of the robot as simple as possible. That is why we are using ready-available parts as much as we can.
  3. Order PCB for daughter board. Daughter board is required to interface the on-board computer with sensors and actuators such as range finders and motor controllers. There are a lot of online shops providing PCB manufacturing services. Selection depends heavily on where you are. Since we are in Germany, we are currently using services from PCB-POOL.
  4. Solder parts. We develop our own daughter board with hobby makers in mind, i.e. without SMD soldering needs, etc. So it should not be a big problem to solder required parts to PCB. In addition, it is necessary to solder some cables and connectors.
  5. Assemble everything together. Here 3D printed parts should be assembled and mounted on the chassis. Parts are easily assembled together. The only required tool is screw driver.
  6. Install software. There are two groups of software components we provide - those running on on-board computer and those running on control computer (such as for example notebook). The referenced wiki article provides detailed description how to install them. All required software is available from our git repository.
  7. After everything is up and running, you might be interested to check out the set of examples we developed. They are inspired by the great online class "Artificial Intelligence for Robotics (cs373) - Programming A Robotic Car" taught by professor Sebastian Thrun. We modified the source code for some algorithms and topics explained in this class to control our robot. We keep the core part unchanged but substitute the control commands with our function calls. As a result, the same algorithm can now control the robot. The following examples are available:
    • Unit1 - one-dimensional localization using histogram filter.
    • Unit2 - speed and position estimation with Kalman filter.
    • Unit3 - position estimation using particle filter.
    • Unit5 - motion control with PID-controller and Twiddle algorithm to find PID parameters.
Tweets by @veterobot