Behaviour commands: in the ROS 2 simulator, and standing still

Behaviour policies now run in the robot's own control software in simulation, and a retrained policy stands still when no speed is commanded.

• The robot's controller now takes behaviour policies: 46 inputs per step, behaviour commands from the driving page and a gait clock kept by the controller; earlier policies run as before
• Seen there: with the default behaviour and no speed command, the behaviour policies stepped in place (summed joint speed 32.9 rad/s, as much as walking)
• Retrained with a zero speed command in a fifth of the default-walk segments: 0.23 rad/s in the simulator; in training physics 0.00 against 1.79 touchdowns per second, drift 0.03 against 0.50 m
• Still open: standing robots tuck a front leg to its joint limit in training physics, 14 of 30 times over 20 s
• Hard terrain unchanged by the retraining: 4 cm stairs 50 against 47 passes of 100