Abstract
Force control of legs of evolving walking machines is
assumed to be essential
in natural soft and
uneven terrain. The main duties of a leg are to support
and propel the vehicle
in co-operation
with the other legs. Force control fills the support
requirement also when a
leg or many legs
penetrate the terrain. It also prevents rolling of the
body due to lift-off or
placement of legs.
Due to the relatively large mass and inertia of the body,
dynamic equations of
a free object are a
basis for "computed-torque" based calculations of desired
body forces to move
or participate in the
applications as desired. The large and varying amount of
friction in the leg
mechanisms while the
body is carried and the practical delays and saturation
in the hydraulic system
increase the
complexity of the process to be controlled.
The main contributions in this thesis are as follows. The
body forces are
transformed to the
supporting legs in two phases: sets of minimum forces
perpendicular to the
resultant body force
and forces parallel to the same resultant. This method
minimizes the
possibility of slippage with
walking machines, where the desired body force is often
close to vertical due
to the weight of the
body.
A load adaptive PI force control method for the hydraulic
actuation system of
MECANT I
consisting of an asymmetric cylinder and a symmetrical
valve has been
developed. The I term of
the controller is changed according to the desired load.
A rule based altitude controller and a dead-zone and
saturation based attitude
controller have
been designed. The first tests with force controlled
vertical actuators show
the usability of the
method and fast responses to deviations in body
orientation. The deviations are
usually corrected
within 1 second.
Original language | English |
---|---|
Qualification | Doctor Degree |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 10 Jun 1994 |
Place of Publication | Espoo |
Publisher | |
Print ISBNs | 951-38-4417-X |
Publication status | Published - 1994 |
MoE publication type | G4 Doctoral dissertation (monograph) |
Keywords
- walking machines
- robots
- force
- control
- motion
- legs
- automation
- hydraulic actuators
- modelling
- simulation