(abstract)
Walking machines have higher potentials than the wheeled machines
for rough terrain traversing. They can isolate the body from
terrain irregularities, avoid undesirable footholds, regulate
their stability and achieve energy efficiency. To fulfill these
potentials, walking machines are required to move in a
terrain-adaptive way. However, the current motion planning method
cannot effectively move the machine so that it can be adaptive to
the terrain. The shortage of motion planning method greatly
lagged the development of walking machine.
In this work, the kinematic aspects of terrain-adaptive
locomotion of walking machines are studied with the purpose to
improve their mobility on rough terrain. The study deals with
two levels of motion planning of walking machines: the gait
planning and the body trajectory planning. Both planning methods
were developed based on the kinematic interaction between the
machine and the terrain.
In the study of gait planning, a free gait algorithm with a new
mode of gait generation, called is developed. The primary gait is
a dominating gait that can be used for most of the walking
conditions, while the secondary gait is used only when the vehicle
cannot move with the primary gait. Four constraints were
elaborated to narrow the searching scopes for gait parameters.
With the proposed primary/secondary gait, the number of possible
choices is greatly decreased and the efficiency of gait generation
will then be improved. By introducing a generalized reference
coordinate (GRC) system, the primary/secondary mode is extended
to the gait planning of both the straight line and the circular
motion. With the GRC system, the constraints evaluation can be
conducted for straight line and circular motion with the same set
of equations. Therefore the trajectory of a walking machine can be
designed more naturally.
In the study of path planning, a new method of path-planning is
proposed based on the terrain evaluation. It is recognized that a
successful movement of the machine is dependent on both the
terrain as well as the machine itself. To generate an optimal
path, it is required to take both aspects into consideration. In
the proposed method, a measure named is introduced to combined the
terrain complexity and the machine mobility together in the course
of path planning. The terrain accessibility is an indicator of the
interdependence of terrain and machine. The greater the terrain
accessibility, the greater the possibility of traversing the
terrain by the machine. The accessibility is obtained based on two
kinds of terrain information: the number of possible foothold and
its distribution in an area considered with respect to the machine
features. With the evaluated accessibility, the terrain can be
reconstructed into a potential field. By search points with
maximum accessibility in this field, an optimal path can be
generated. Such a path could offer the best opportunity to place
machines' feet moving by a certain gait over a rough terrain.
Parallel to the development of motion planning method, some
biological study on the behavior of animal locomotion on rough
terrain were conducted. In the experiment the movement of a
cockroach was observed using high speed camera system. The results
allow us to understand better the behavior and motion strategy
of animals and help us to improve the in terrain-adaptive motion
planning of walking machines.
As a conclusion of this research, it is suggested that the
motion planning of walking can be carried out through the approach
of kinematic interaction. It is the machine/terrain interaction
that affects the terrain-adaptive locomotion. An
interaction-based motion control will be an effective approach in
development of autonomous walking machines.