MIT News breed · Science

MIT Researchers Develop Collision-Free Autonomous Flight System

MIT researchers have created a new autonomous navigation system, named SANDO, that guarantees collision-free flight paths for uncrewed aerial vehicles (UAVs) even in unknown and dynamic environments.

The SANDO system plans efficient flight paths that mathematically avoid collisions with unpredictable moving obstacles, a significant improvement over systems that require static or known environments.

This technology is particularly useful for high-risk applications such as search and rescue in collapsed buildings, mine exploration, and package delivery in crowded areas.

Unlike existing planners, SANDO creates time-sensitive safety corridors by estimating the future movements of obstacles based on their maximum velocity, ensuring safety even as the environment changes.

The system uses a heat-map approach to identify and avoid dangerous areas, guiding the UAV along a more efficient route.

Once a safe corridor is established, SANDO optimizes the trajectory for speed and continuously adjusts the path as the UAV travels, reacting quickly to changes.

In simulations and real-world tests with a UAV, SANDO successfully avoided all collisions and reached its goal faster than several other advanced systems.

AI-samenvatting op basis van de bron.