Please enable JavaScript.
Coggle requires JavaScript to display documents.
Minimizing the time rescue robots spend scanning and learning an…
Minimizing the time rescue robots spend scanning and learning an environment
Edge Computing
Edge Devices
Edge computing is a networking philosophy focused on bringing computing as close to the source of data as possible in order to reduce latency and bandwidth use (i.e. fewer processes in cloud, more processes on personal computer, IOT device or edge server).
Edge servers
edge nodes
edge routers
VSLAM
Tracking
The process a robot uses to map and navigate through an environment using only visual images provided by the camera. The final map is generated dynamically through the feature points which are basically the markers in each and in the combination of the images that map the environment.
Keyframe selection
global map optimization
The ability to dynamically update or map-out changes to the environment across all robots present.
Aim is to improve map accuracy simultaneously
sensor fusion
combining and using data of all sensors in order for:
3D mapping with just the data needed
makes the processing quicker and simpler since all the data is gathered and guided already
bundle adjustment
technique used in computer vision to refine the alignment of 2D images and their corresponding 3D points for more accurate 3D reconstruction.
robot's trajectory (poses)
3D positions of the observed landmarks (features)
visual information (camera-->keypoints/frames)
dynamic mapping
and navigation
indoors-->computer vision
odometry sensor
cameras
outdoors--> GPS signals
computer vision
AI that uses digital images and models in order to identify and classify objects, while also reacting to them
Seamless identification of both static or dynamic objects/keypoints/targets/etc in environment