# Skydio Drone Flight to a Landing Pad in MuJoCo > Drift builds a Skydio X2 simulation where a drone uses LiDAR to navigate around obstacles, reach a target table, and land on a marked pad. **Author:** Drift Team **Published:** 2026-09-12 **Tag:** Product **Reading time:** 3 min **Canonical URL:** https://godrift.ai/blogs/skydio-drone-flight-to-a-landing-pad ---## **Setting Up the Drone Mission** Flying from one table to another sounds simple until obstacles get in the way. For this simulation, we start with an indoor room containing a few obstacles and two tables placed across from each other. The Skydio X2 starts at rest on one table, while a landing pad marks the destination on the other. Drift uses the existing X2 model and adds a LiDAR sensor to help the drone detect obstacles during the flight. This builds on the same MuJoCo workflow we used in our earlier[ autonomous Skydio X2 flight simulation](https://www.godrift.ai/blogs/drone-flight-simulation-mujoco), but adds a more constrained environment and obstacle avoidance to the mission.
## **From Takeoff to Landing** The task follows a simple sequence: **home position → takeoff → obstacle avoidance → target table → landing** The X2 takes off from its starting position and navigates through the room while accounting for the obstacles along its path. Once it reaches the destination, the drone aligns itself with the landing pad and performs a controlled landing. The entire flight runs autonomously, with no manual piloting. ## **What the Drone Has to Solve** The flight combines several problems into one task. The LiDAR provides information about the surrounding environment. The navigation logic uses that information to avoid obstacles and move toward the destination, while the final approach requires the drone to control its position accurately enough to land on the pad. It's similar to the navigation challenge in[ our MuJoCo maze simulation](https://www.godrift.ai/blogs/maze-navigation-mujoco), where a robot has to reach a goal while dealing with obstacles along the way. The difference here is that the X2 has to solve the problem in three dimensions. Simulation makes it possible to test this kind of behavior safely before putting a real drone through the same flight. As we covered in[ Why Robots Practice in Simulation Before the Real World](https://www.godrift.ai/blogs/why-robots-train-in-simulation), virtual environments provide a place to test, fail, and iterate without risking hardware. ## **From Simple Flight to Autonomous Missions** A flight between two tables is a small task, but it brings together perception, navigation, obstacle avoidance, and precise landing. From here, the same setup could be extended with more complex environments, dynamic obstacles, or longer autonomous missions. That's where drone simulation becomes more interesting: start with a controlled flight, then keep adding the problems a real autonomous system would have to solve. ## **FAQ** ### **What drone is used in the simulation?** The simulation uses the Skydio X2 model in MuJoCo. ### **What sensor does the drone use?** A LiDAR sensor is added to help the X2 detect obstacles during its flight. ### **What is the goal of the task?** The drone must autonomously fly from its starting table to a landing pad on another table while avoiding obstacles. ### **What does this simulation demonstrate?** It combines perception, navigation, obstacle avoidance, and precise landing into one autonomous drone task. ## **Related reading** - [Simulating Autonomous Drone Flight With the Skydio X2 in MuJoCo](https://www.godrift.ai/blogs/drone-flight-simulation-mujoco) - [Watch a Robot Solve a Maze on Its Own in MuJoCo](https://www.godrift.ai/blogs/maze-navigation-mujoco) - [Why Robots Practice in Simulation Before the Real World?](https://www.godrift.ai/blogs/why-robots-train-in-simulation) - [MuJoCo: The Physics Engine Robotics Research Quietly Runs On](https://www.godrift.ai/blogs/what-is-mujoco)