Skippy Project Videos and Animations
This page collects together some of the videos and animations we have
made. Some animations are available both as a video for you to play
directly and as a Matlab .mat file that you can play using the
function showmotion in Spatial v2, which will give you a much better viewing
experience. To download the .mat file, click where you see the
icon SV2.
Skippy Balancing, Hopping and Falling Over
(Oct 2023)
This video, prepared by Antony, shows Skippy balancing and hopping in a
vertical plane, and falling over a lot. A shoe in the shape of a blade
is fitted to Skippy's foot, so that Skippy is effectively balancing on a knife
edge and can therefore use a planar balance controller. The video first
shows Skippy balancing, making a small hop in place, recovering its balance,
and making another hop in place. It then shows Skippy making a two-hop
sequence in which the first hop is in place and the second is a travelling
hop. The rest of the video then shows Skippy falling over again and
again. This last part shows how robust Skippy really is: the machine
has now fallen many dozens of times without ever sustaining any
damage. Everything that Skippy does in this video is accomplished using
a single actuator.
Making a Double Backflip Using a Springy Leg
(June 2023)
This animation, prepared by Juan, shows a planar double pendulum with a
springy leg get up off the floor, balance, crouch, launch into a double
backflip, control its motion in flight, land, absorb the energy of landing,
and come to rest in a balanced position—all of it accomplished using
only a single actuated joint. It was the climax of Juan's
Ph.D. studies, which focussed on the problem of balancing and hopping with a
springy leg, including techniques to absorb energy from the spring on landing,
and to damp out vibrations in the spring while balancing. It is a clear
improvement on Morteza's original work shown below. For more details on
Juan's work on this topic see publications 21, 23, 27 and 31.
Control of Absolute Motion While Balancing
(Dec 2021)
This animation shows a planar triple pendulum balancing on a knife edge while
making its tip follow a trajectory shown in green. The novelty here is
that the robot is controlling the absolute motion of its tip, using an
operational-space motion controller, and using the null space of the motion
task to balance; but both the operational space and the null space involve the
orientation of the bottom link about the knife edge, which is the robot's
passive degree of freedom, so the robot has both an underactuated operational
space and an underactuated null space. This animation is an excerpt
from Roy's presentation at ICAR 2021. To see the complete presentation
click here; and for more details
see publication 26.
Tippy Balancing in 2D Using the Crossbar
(Feb 2018)
This video shows Tippy balancing while simultaneously following a command
signal that specifies the motion of the crossbar. You can see the
command signal in the left part of the video. The balance controller
includes a balance offset observer that measures the difference between true
and estimated balanced configurations so that the controller can improve its
estimate. You can see this happening during the first few
seconds, and again near the end of the video when the robot is being
pushed. The command signal asks Tippy to make some very fast movements,
and you can see that Tippy responds very quickly. You can also see it
leaning in anticipation of the command signal. Finally, the video shows
Tippy resisting and recovering from large external disturbances.
For more details see publication 17.
Bend-Swivel Balance Control in 3D
left: (2012) SV2,
right: (2016) SV2
These animations show what bend-swivel balance control looks like. In
this strategy, the bend angle is controlled by a 2D balance controller, and
the task of the swivel controller is to keep the bend plane vertical and
control the overall heading of the robot. The animation on the left
shows Morteza's original implementation; and the one on the right is a more
recent implementation by Roy on a Tippy-like robot. Both keep their
balance very well, but neither tracks motion commands accurately while
swivelling. The likely reason is gyroscopic forces. See
publications 6 and 8.
Leaning in Anticipation
left: (2015) SV2,
right: (2016) SV2
The one on the left was Roy's first demonstration of leaning in
anticipation. It was presented at talk 1, and shows a simple bending
and straightening movement performed twice without leaning in anticipation,
and then twice more with leaning in anticipation. The graph of the
response is here. The one on the
right begins with a sequence of movements performed without leaning in
anticipation, and continues with a similar sequence performed with leaning in
anticipation and much higher feedback gains, resulting in much faster, sharper
movements. Gains this high would be completely impractical without
leaning in anticipation. See publication 14.
Balancing in Combination with Other Motions
(2015) SV2
This is the animation that Roy showed at Int. Symp. Robotics Research (see
publication 10), and again in talk 1. It is the only animation that
shows the balance controller operating together with a joint-space motion
controller. It is also the only animation to show a branched kinematic
tree; and it is the last animation that Roy made before working out how to do
leaning in anticipation. The animation begins with three bending and
straightening motions. In the first, the balance controller operates
the lower joint; in the second it operates the upper two joints; and in the
third it operates all three. In the remainder of the animation, the
balance controller operates only the lower joint.
Morteza's Great Leaps Forward
(2012)
The animation on the left shows Morteza's balance controller performing single
hops beginning and ending in a balanced upright configuration. The
robot is based on the Acrobot, as
defined here. The sequence of actions is more complicated
than it appears. First, the robot must maintain its balance in the
initial position. Then it must crouch down and lean forward
(deliberately losing its balance), and then push with the right speed and
timing in order to reach the right lift-off momentum (linear and angular) as
the foot leaves the ground. Then it has to control its foot
trajectory during flight in order to land on the right spot. Finally,
it has to recover its balance after landing, and return to an upright
configuration. And all of this is done using just one actuator.
See publications 4 and 6. The animation on the right, which Morteza
never published, shows the same movement, using the same control system, but
on a robot with a springy leg. It shows that the balance controller can
cope with the large disturbances caused by a spring that the control system
doesn't know is there. Although never published, this was a pioneering
result.