Design and Implementation of Soft-I-Robot for Face Detection and Obstacle Avoidance Using Neural Networks and Genetic Algorithm

The Internet-based security Soft-i-Robot is modeled using Soft computing paradigms for problem solving and decision-making in complex and ill-structured situations. Soft-i-Robot monitors the workspace with multimedia devices and sensor using an Internet application program. The model has sensory subsystems such as Intruder detection which, detects intruder, captures image and sends to server, and an Obstacle Avoidance Unit to detect the objects in the path of the mobile robot. These multiple features with hybrid Soft computing techniques depart the developed Soft-i-Robot from the existing developments, proving that the streaming technology-based approach greatly improves the sensibility of robot tele-operation. The relatively powerful online robots available today provoke the simple question, in terms of two competing goals: recognition accuracy and computing time. Improved recognition accuracy and reduced computing time for face recognition of the intruder is obtained using Morphological Shared Weight Neural Network. To obtain a collision-free optimized path, Soft-i-Robot uses derivative free Genetic Algorithm. With rapid expansion of Robotics and Soft computing paradigms, robotic technology touches upon self-understanding of humans, socio-economic, legal and ethical issues leading to improved performance rate and information processing capabilities