Abstract: This paper presents a method for steering velocity bounded mobile robots in environments with partially known stationary obstacles. The exact location of obstacles is unknown and only a probability distribution associated with the location of the obstacles is known. Kinematic model of a 2-wheeled differential drive robot is used as the model of mobile robot. The presented control strategy uses the Artificial Potential Field (APF) method for devising a desired direction of movement for the robot at each instant of time while the Constrained Directions Control (CDC) uses the generated direction to produce the control signals required for steering the robot. The location of each obstacle is considered to be the mean value of the 2D probability distribution and similarly, the magnitude of the electric charge in the APF is set as the trace of covariance matrix of the location probability distribution. The method not only captures the challenges of planning the path (i.e. probabilistic nature of the location of unknown obstacles), but it also addresses the output saturation which is considered to be an important issue from the control perspective. Moreover, velocity of the robot can be controlled during the steering. For example, the velocity of robot can be reduced in close vicinity of obstacles and target to ensure safety. Finally, the control strategy is simulated for different scenarios to show how the method can be put into practice.
Abstract: A topologically oriented neural network is very
efficient for real-time path planning for a mobile robot in changing
environments. When using a recurrent neural network for this
purpose and with the combination of the partial differential equation
of heat transfer and the distributed potential concept of the network,
the problem of obstacle avoidance of trajectory planning for a
moving robot can be efficiently solved. The related dimensional
network represents the state variables and the topology of the robot's
working space. In this paper two approaches to problem solution are
proposed. The first approach relies on the potential distribution of
attraction distributed around the moving target, acting as a unique
local extreme in the net, with the gradient of the state variables
directing the current flow toward the source of the potential heat. The
second approach considers two attractive and repulsive potential
sources to decrease the time of potential distribution. Computer
simulations have been carried out to interrogate the performance of
the proposed approaches.