Embodied AI agents have a physical or simulated body through which they perceive their surroundings via sensors and act on them through movement. Examples include warehouse and household robots, humanoid robots and characters in simulations where robots are trained before real-world use. The challenge lies in combining perception, planning and safe action in an unpredictable world.
Collaborative robots, or cobots, are industrial robots designed to work directly alongside people without a separating safety fence. Sensors, limited forces and speeds, and safety functions ensure that they stop or move away on contact; requirements are set out in standards such as the ISO 10218 series. Cobots handle tasks such as loading machines, screwing or packing, while employees take care of inspection and special cases.
An intelligent agent is a system that perceives its environment, processes information and chooses actions on its own in order to reach a goal. Today the term usually refers to software built on large language models that follows instructions, draws on knowledge and calls tools such as a search function or an API. How much an agent may do independently should be decided deliberately. Example: an agent answers customer questions using the product documentation.
Autonomous vehicles move fully or partly without human intervention, using cameras, radar, lidar and AI software. The SAE J3016 standard defines six levels – from 0 (no automation) to 5 (fully driverless everywhere). AI handles perception, predicting the behaviour of other road users and route planning. Besides cars, this covers robotaxis, lorries, shuttles and warehouse vehicles; approval and liability are regulated separately in each country.
