Fuzzy neural networks for obstacle pattern recognition and collision avoidance of fish robots.

Soft Computing (Impact Factor: 1.3). 01/2008; 12:715-720. DOI: 10.1007/s00500-007-0245-0
Source: DBLP

ABSTRACT The problems of detection and pattern recognition of obstacles are the most important concerns for fish robots’ path planning
to make natural and smooth movements as well as to avoid collision. We can get better control results of fish robot trajectories
if we obtain more information in detail about obstacle shapes. The method employing only simple distance measuring IR sensors
without cameras and image processing is proposed. The capability of a fish robot to recognize the features of an obstacle
to avoid collision is improved using neuro-fuzzy inferences. Approaching angles of the fish robot to an obstacle as well as
the evident features such as obstacles’ sizes and shape angles are obtained through neural network training algorithms based
on the scanned data. Experimental results show the successful path control of the fish robot without hitting on obstacles.

  • [Show abstract] [Hide abstract]
    ABSTRACT: Special types of mobile robots are necessary for the pipeline monitoring tasks. One of the bio-mimetic robot applications serves the purpose of exploring pipelines, spotting any troubled areas or malfunctions and reporting its data. Moreover, the robot is able to prepare for and react to any abnormal routes in the pipeline. In order to move effectively along a pipeline, the robot's movement will resemble that of insects. When situated in massive pipelines with complex routes, the robot should have the abilities of route recognition and mobility while monitoring the pipelines. The robots have grippers to hold the pipes and four arms with three DOF joints to move on the pipelines. This monitoring task is to prevent a major system failure by preemptively recognizing any minor or partial malfunctions.
    Computer and Automation Engineering (ICCAE), 2010 The 2nd International Conference on; 03/2010
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Complex structured environments offer fish advantages as places of refuge and areas of greater potential prey densities, but maneuvering through these environments is a navigational challenge. To successfully navigate complex habitats, fish must have sensory input relaying information about the proximity and size of obstacles. We investigated the role of the pectoral fins as mechanosensors in bluegill sunfish swimming through obstacle courses under different sensory deprivation and flow speed conditions. Sensory deprivation was accomplished by filming in the dark to remove visual input and/or temporarily blocking lateral line input via immersion in cobalt chloride. Fish used their pectoral fins to touch obstacles as they swam slowly past them under all conditions. Loss of visual and/or lateral line sensory input resulted in an increased number of fin taps and shorter tap durations while traversing the course. Propulsive pectoral fin strokes were made in open areas between obstacle posts and fish did not use the pectoral fins to push off or change heading. Bending of the flexible pectoral fin rays may initiate an afferent sensory input, which could be an important part of the proprioceptive feedback system needed to navigate complex environments. This behavioral evidence suggests that it is possible for unspecialized pectoral fins to act in both a sensory and a propulsive capacity.
    Journal of Experimental Biology 08/2013; 216(Pt 16):3084-9. · 3.24 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Adaptive neurofuzzy inference systems (ANFIS) represent an efficient technique for the solution of function approximation problems. When numerical samples are available in this regard, the synthesis of ANFIS networks can be carried out exploiting clustering algorithms. Starting from a hyperplane clustering synthesis in the joint input-output space, a computationally efficient optimization of ANFIS networks is proposed in this paper. It is based on a hierarchical constructive procedure, by which the number of rules is progressively increased and the optimal one is automatically determined on the basis of learning theory in order to maximize the generalization capability of the resulting ANFIS network. Extensive computer simulations prove the validity of the proposed algorithm and show a favorable comparison with other well-established techniques.
    Advances in Fuzzy Systems 01/2012; 2012:1-12.


Available from