Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Robotic Simualtion for LiDAR-based Forest Mapping with UAVs
The main objective of the project is to develop a demo for forest mapping with a quadrotor equipped with LIDAR/cameras in photorealistic simulation environment.
Aerial robots equipped with LiDAR and cameras may provide a repeatable, fast, and accurate solution to enhance data collection for mapping and modelling of forests from above and below the canopy. In this project we want to develop a solution for LiDAR- or vision-based 3D mapping of a forest in a photorealistic simulator, leveraging a UAV that can fly and acquire useful data within the surrounding environment. First, the student will identify the best option to render a forestry ecosystem and to simulate LiDAR/camera sensors to acquire point clouds of the environment (e.g., Flightmare, Unreal, etc.). Second, the student will include a quadrotor model embodied with autonomously flight capabilities to move around the scene in a controlled and repeatable way. Third, the student will incorporate the aforementioned sensors on the drone and develop a demo to acquire 3D point clouds of the surrounding while performing a desired trajectory.
Aerial robots equipped with LiDAR and cameras may provide a repeatable, fast, and accurate solution to enhance data collection for mapping and modelling of forests from above and below the canopy. In this project we want to develop a solution for LiDAR- or vision-based 3D mapping of a forest in a photorealistic simulator, leveraging a UAV that can fly and acquire useful data within the surrounding environment. First, the student will identify the best option to render a forestry ecosystem and to simulate LiDAR/camera sensors to acquire point clouds of the environment (e.g., Flightmare, Unreal, etc.). Second, the student will include a quadrotor model embodied with autonomously flight capabilities to move around the scene in a controlled and repeatable way. Third, the student will incorporate the aforementioned sensors on the drone and develop a demo to acquire 3D point clouds of the surrounding while performing a desired trajectory.
**Work packages:**
- Familiarization with state-of-the-art literature on LiDAR- and camera-based forest mapping
- Development of the photorealistic forest in the simulation environment
- Integration of LiDAR and camera sensor models
- Integration of a UAV with autonomous flight capabilities
- Development of a repeatable demo for point clouds acquisition during flight
- Simulation experiments for 3D reconstruction from sensor data
**Requirements:**
- Highly motivated and interested in robotic, drones and simulation
- Methodological and goal-oriented working behavior
- Strong background in the following areas: UAV, mapping, robotic simulation (experience with LiDAR and camera sensors is surely advantageous)
- Excellent programming skills (C++, Python)
- Knowledge of Flightmare, Unity or Unreal Engine is a big plus
**Work packages:**
- Familiarization with state-of-the-art literature on LiDAR- and camera-based forest mapping
- Development of the photorealistic forest in the simulation environment
- Integration of LiDAR and camera sensor models
- Integration of a UAV with autonomous flight capabilities
- Development of a repeatable demo for point clouds acquisition during flight
- Simulation experiments for 3D reconstruction from sensor data
**Requirements:**
- Highly motivated and interested in robotic, drones and simulation
- Methodological and goal-oriented working behavior
- Strong background in the following areas: UAV, mapping, robotic simulation (experience with LiDAR and camera sensors is surely advantageous)
- Excellent programming skills (C++, Python)
- Knowledge of Flightmare, Unity or Unreal Engine is a big plus