No abstract

In this paper, we propose a genetic algorithm based approach to determine the pose of an object in Automated Visual Inspection having three degrees of freedom. We have investigated the effect of noise at 20 dB SNR and also mismatch resulting from incorrect correspondences between the object space points and the image space points, on the estimation of pose parameters. The maximum error in translation parameters is less than 0.45 cm and rotational error is less than 0.2 degree at 20 dB SNR. The error in parameter estimation is insignificant upto 7 pairs of mismatched points out of 24 points in object space and the results skyrockets when 8 or more pairs of points are mismatched. We have compared our result with that obtained by least square technique and it shows that GA based method outperform the gradient based technique when the number of vertices of the object to be inspected is small. These results have clearly established the robustness of GA in estimating the pose of an object with small number of vertices in automated visual inspection.

Unmanned Aerial Vehicles (UAVs) have been developed from remotely guided drones into experimental Unmanned Combat Aerial Vehicles (UCAVs) of today for combating terrorism. This technology was pioneered by the military and then expanded into civilian uses, e.g. Film production. The growth was endorsed by the advancement of computing and navigation technology, which leads to an important research issue in the design and development of Autonomous UAVs. Recently, the demands for small platform of autonomous UAV have been increasing for urban operations, which require a mix of small, distributed air and ground based sensors to detect and track small, distributed targets in the urban clutter. However, current UAVs have numerous potential failure points. Doing an experiment to improve UAV system reliability is sometimes difficult due to many restrictions and limitations that prevent it from being done in the real world. Simulation is therefore chosen as the approach of this project to study the autonomous behaviors of mini UAVs deployed for urban operation. The study includes: (i) defining the operational requirements of a modeled UAV deployed in urban environment, (ii) testing the conceptual design and operation of modeled UAV in a virtual environment using simulation platform, and (iii) refining the conceptual design and operation based on new defined operational requirements. The development of the simulation platform is divided into two stages. The first stage is the formulation of a conceptual design for the mini UAV named RRCUAV, with an emphasis on the maneuverability and adaptability for urban operation. The second stage is the establishment of a UAV simulator, which is the product between the RSuav module that focuses on aerodynamic models and intelligent system, and ROBOSIM plug-ins, which specialize in 3D graphical drawing and the creation of virtual environment. With the UAV simulator created, conceptual operation is iteratively improved and might be followed by redesigning the mini UAV to accommodate the requirements. A set of autonomous behaviors are developed and they include: (i) Automatic Takeoff, (ii) Go-to-waypoint operation followed by Automatic Ground Tracking (iii) Automatic Landing, and (iv) Obstacle Avoidance. The result of study shows that RRCUA V is generally suitable for urban operation because it is able to: (i) perform takeoff and landing at reasonable space, accuracy, and time, (ii) perform stable go-to-waypoint operation at limited waypoint separation distance, (iii) provide stable surveillance operation, (iv) provide safety procedure in the event of power shortage and loss of communication, and (v) perform obstacle avoidance over business district with nearly 100% robustness. v