Researchers have developed a hexapod robot able to adapt after losing a leg by adjusting its set of behaviors to suit its new design.
The hexapod robot was designed to refer to a self-model, which allows it to understand how it is ideally supposed to work and recognize when it becomes less efficient. If the robot suffers damage that impacts its performance, such as losing a leg, it will automatically search for ways to improve its efficiency. In contrast to previous theories, which called for the robot analyze its new gait and diagnose the damage to compensate, the research team designed the robot to generate different gaits, test them and then compare them with reality.
The researchers compared the robot’s adaptability with the human method of instinctively avoiding pain, even if they do not fully understand the injury that is causing the pain.