Securing the interface: Safety-critical interaction between humans and mobile robots

Forty years ago, if you wanted to park your car, you had to control the speed, acceleration and manoeuvring yourself. Today, you can buy a car that parks itself. Forty years ago, if you wanted to remove dust from your house, you had to move around it with a suction device, pointing the opening here and there. Today, you turn a flat round object on and go make a cup of coffee. Forty years ago, if you wanted some behavioral-safety principles for robots, there were Asimov´s laws. Today, if you want some behavioral-safety principles, there are Asimov´s laws. (And some health-and-safety laws requiring interlocks to prevent people being within the reach-radius of fixed-base industrial robots. And there are people working on principles of engagement for artifical warriors, which we may imagine are not likely to cover the full range of robotic safety needs.) To my mind, the principles have not kept pace with the practice. I want to make a start on catching up.