Show-off

Show-off of our platform of cases that illustrate all systems working

Collision avoidance

Collision avoidance is a crucial feature in any drone swarm system, and in the Swarm in Block is no exception. With collision avoidance technology embedded in the system's code, it ensures the safety and efficiency of the drone swarm while performing various tasks.

Collision avoidance technology in drones works by using various sensors, such as ultrasonic and infrared sensors, to detect obstacles and other drones in their path. In our case, we use visual odometry with ArUco Maps so that each drone knows where it is. This concept is based in the OpenCV analysis of the video camera under the drone; knowing, by this way, where it is comparing with a pre-installed map in the system.

One of the primary advantages of the avoidance feature is its ability to operate in dynamic environments. The system can continuously monitor the environment and adjust the swarm's path in real-time to avoid collisions. This feature makes it ideal for various applications, such as agricultural monitoring or package delivery, where drones need to navigate through a changing environment.

ArUco Map grid

The Swarm in Block uses ArUco maps for the visual odometry of each drone in the swarm. This decision was made because Aruco maps offer several advantages over other visual odometry methods, including increased safety, greater control of environments, and portability.

This portability is especially useful in applications where the drone swarm needs to be deployed in different environments or locations. This make the swarm be able to rise fly in any place; creating, by consequence, a better control of flies, because any place can be converted to a base for the swarm.

Aruco Map in real life

Note that the green tape we used worked better in the light conditions of our room, that way we got better results wit the Clovers cameras in terms of precision.

Show-off

To show off the collision avoidance feature of the Swarm in Block, two videos demonstrate the system's capabilities.

In one video, we can see a drone swarm navigating until the system check that exist a possible collision to happens. Then it stops, avoiding the shock in same plane with the another drone, letting it pass.

A parallel colision avoided

In another video, we can see the system operating also in a dynamic environment, adjusting the vertical swarm's path in real-time to avoid the frontal collision between the drones.

Parallel collision avoided

Coordenated Real Fly

As part of our ongoing efforts to improve the Swarm In blocks, we are preparing to take the next step in our development process: coordinated real fly testing with collision avoidance code and Aruco map odometry logic in its full complexity. Achiving marks as geometrical forms (circle, square, triangle, polygons at all).

Case 1: Swarm with LED

The drones are synchronized in the swarm also with LEDs, being able to construct simple to complex LED formations. This case proves the work of the swarm in a controlled enviroment where we are not susceptible to odometry or control problems, as we faced due to hardware and firmware complications. Besides that, Clover LEDs are really beautifull when combined.

Simple example of synchronized LEDs

Another example of synchronized LEDs

Case 2: Swarm Takeoff

When we get to the real world, the hardware and firmware issues can be really frustating. That was our case for the past few months. Even with that, we overcome our main problems in order to make a swarm takeoff coordenated between this two Clovers. Unfortunatelly, we could not make more demos at the moment, but soon we work on this obstacles, new show offs will be displayed.

Coordenated Takeoff

This show off demonstrate and validate that the platform it is functional and it is closed to be finished.

In conclusion

Collision avoidance is an essential feature in any drone swarm system, and the Swarm in Block implementation of this technology ensures the safety and efficiency of the drone swarm while performing the mobility. The videos of the collision avoidance show off its precision and effectiveness, providing a clear picture of how collision avoidance technology works in drone swarms.

With that in mind, with the odometry solution that bests fits Clover, the Aruco Map, we were able to perform Real life test to Show our platform. As soon as we get more results we will show in this section, so stay tuned!