Field robotics

Robots outside the lab: under water, on the water, in the air and outdoors on land. Each setting breaks something that works indoors, from cameras that see a few metres under water to drones that carry 27 g of everything.

Concepts

• Underwater perception: light, navigation and reconstruction under water.
• Aerial swarms: small drones, spoofing and safe flight.
• Singapore scenarios: sea, air and land around Singapore.
• Collaborative SLAM, filed under multi-robot systems.

Entities

Datasets, platforms, tools and groups that come up in this subfield: BlueROV2, Doppler velocity logs, Spot, Jackal, Husky, Unitree robots, Jetson. The entity index has the rest.

Open questions

• How does a vehicle know its pose has gone wrong when the water fools its confidence estimates as well?
• What is worth sending over an acoustic modem that carries kilobits per second?
• How much of a ground robot’s perception fits on a 27 g drone?

Papers

Other subfields: perception, mapping and localisation, semantic mapping, planning and navigation, control, robot learning, multi-robot systems, safety and verification, security, Human-robot interaction.