sensor fusion for layman: learning to ride a bike is learning to fuse information from visual cortex and vestibular to keep :

• balance
• while keeping the bike away from traffic and static obstacles

Sensor fusion for engineers: the branch of computer science and associated algorithms used to combined multiple sensor data to reduce uncertainty inherent to each sensor output. Intro There are 2 dimensions to sensor fusion:

Level (from wikipedia):

• level 0: data alignement
• level 1: Entity assessment (e.g. signal/feature/object)

Type:

• Competitive: dual channel approach ( 2 overlaping cameras)
• Complementary: appending overlapping sensor data (e.g. front and rear camera)
• Cooperative: complemntary to add more incormation (2 cameras to rnablr stereoscopic views)