Related Experiment Video
Updated: Mar 21, 2026
03:49
Evaluating Flight Performance and Eye Movement Patterns Using Virtual Reality Flight Simulator
Published on: May 19, 2023
1.6K
All eyes, no IMU: learning flight attitude from vision alone
Jesse J Hagenaars1, Stein Stroobants1, Sander M Bohté2,3
1MAVLab, Control & Operations department, Faculty of Aerospace Engineering, TU Delft, Delft, The Netherlands.
Npj Robotics
|March 20, 2026
Summary
This study introduces the first vision-only flight control for drones, using an event camera to estimate attitude without inertial sensors. This breakthrough enables insect-scale robots to fly autonomously using only visual input.
Area of Science:
- Robotics
- Computer Vision
- Aerospace Engineering
Background:
- Flying animals rely on vision for attitude control, unlike robots that typically use inertial sensors (accelerometers, gyroscopes).
- Existing autonomous flight systems often require dedicated gravity sensors, limiting their applicability in diverse environments.
Purpose of the Study:
- To develop and demonstrate a vision-only flight control system for quadrotor drones.
- To enable autonomous flight in generic environments without relying on inertial measurement units.
Main Methods:
- Utilized a downward-facing event camera on a quadrotor drone to capture visual data.
- Developed a recurrent convolutional neural network (CNN) trained via supervised learning to estimate attitude and rotation rate from event streams.
- Integrated the vision-based system into the drone's flight control loop, replacing traditional inertial sensors.
Main Results:
- Successfully demonstrated real-world flight control using only an event camera and a low-latency neural network.
- The vision-only system proved capable of replacing the inertial measurement unit in a standard flight control architecture.
- Investigated network generalization across environments, finding that memory and horizon cues improved performance, while narrower fields of view enhanced generalization.
Conclusions:
- Vision-only flight control is a viable and promising approach for autonomous aerial robots.
- This technology can pave the way for insect-scale flying robots with enhanced autonomy.
- The developed system offers a potential alternative to inertial sensors in robotic flight control.
More Related Videos
Related Concept Videos
01:24Vision
61.3K
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
61.3K
01:24Absolute Motion Analysis- General Plane Motion
678
Visualize a drone, with its propellers spinning rapidly, hovering mid-air. The fascinating movements and operations of this drone can be comprehended by applying the principle of general plane motion.
As the drone's propellers rotate, an upward force is generated that counteracts the force of gravity, enabling the drone to lift off from the ground. This initial movement of the drone is along a straight path, representing a form of translational motion. In this phase, every point on the...
As the drone's propellers rotate, an upward force is generated that counteracts the force of gravity, enabling the drone to lift off from the ground. This initial movement of the drone is along a straight path, representing a form of translational motion. In this phase, every point on the...
678
01:12Observational Learning
1.2K
Albert Bandura's observational learning, also known as imitation or modeling, occurs when a person observes and imitates another's behavior. It is a quicker process than operant conditioning. A well-known example is the Bobo doll study, where children who saw an adult acting aggressively towards the doll were more likely to act aggressively when left alone, compared to those who observed a nonaggressive adult. Many psychologists view observational learning as a form of latent learning...
1.2K
01:15Depth Perception and Spatial Vision
2.6K
Depth perception is the ability to perceive objects three-dimensionally. It relies on two types of cues: binocular and monocular. Binocular cues depend on the combination of images from both eyes and how the eyes work together. Since the eyes are in slightly different positions, each eye captures a slightly different image. This disparity between images, known as binocular disparity, helps the brain interpret depth. When the brain compares these images, it determines the distance to an object.
2.6K
02:16Light Acquisition
9.8K
In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.
9.8K
01:22Purposive Learning
570
E. C. Tolman emphasized the purposiveness of behavior — the idea that much of our behavior is goal-directed. For instance, employees who aim for a promotion work diligently to meet their targets. Tolman argued that when classical conditioning and operant conditioning occur, the organism acquires certain expectations. In classical conditioning, a child might fear a dog because they expect it to bite. In operant conditioning, a person might consistently work overtime because they expect a...
570