Drones Learn Autonomous Flying by Imitating Cars and Bikes
The algorithm DroNet allows drones to fly completely by themselves through the streets of a city and in indoor environments. It produces two outputs for each single input image: a steering angle to keep the drone navigating while avoiding obstacles, and a collision probability to let the drone recognize dangerous situations and promptly react to them.
The algorithm learns to solve complex tasks from a set of ‘training examples’ that shows the drone how to do certain things and cope with difficult situations, much like children learn from their parents or teachers. In this case, cars and bikes are the drones’ teachers.
Top Stories
INSIDERManned Systems
Turkey's KAAN Combat Aircraft Completes First Flight - Mobility Engineering...
INSIDERMaterials
FAA Expands Boeing 737 Investigation to Manufacturing and Production Lines -...
INSIDERImaging
New Video Card Enables Supersonic Vision System for NASA's X-59 Demonstrator -...
INSIDERManned Systems
Stratolaunch Approaches Hypersonic Speed in First Powered TA-1 Test Flight -...
INSIDERUnmanned Systems
Army Ends Future Attack and Reconnaissance Helicopter Development Program -...
ArticlesEnergy
Can Solid-State Batteries Commercialize by 2030? - Mobility Engineering...
Webcasts
AR/AI
From Data to Decision: How AI Enhances Warfighter Readiness
Energy
April Battery & Electrification Summit
Manufacturing & Prototyping
Tech Update: 3D Printing for Transportation in 2024
Test & Measurement
Building an Automotive EMC Test Plan
Manufacturing & Prototyping
The Moon and Beyond from a Thermal Perspective
Software
Mastering Software Complexity in Automotive: Is Release Possible...