In recent years, the number of human made objects performing a broad range of tasks in various space missions has increased. Because of limited area in space-crafts, it is essential to keep every tool compact. One of the advanced types of space-based robots is a reconfigurable manipulator. A reconfigurable manipulator is capable of changing its kinematics without the use of any modular joint. In this study, a reconfigurable manipulator with two telescopic and six revolute joints and floating base has been considered. The advantage of this manipulator is its minimal weight and high mobility capable of operating in various configurations. The position of each link can be defined by homogenous transformation matrices and D-H parameters. Using the information from forward kinematics, the inverse kinematics can also be conducted to identify joint angles for the specific manipulator's end effector position and orientation in 3D space considering the redundancy of the reconfigurable manipulator. The nonlinear optimization methodology based on Sequential Quadratic Programming (SQP) technique has been utilized to identify the joints' variables with the singularity and obstacle avoidance for the given end-effector position and orientation. Effective control laws have been proposed to perform a large number of tasks while the manipulator is floating in the space. Proportional, Integral, and Derivative (PID) controllers are utilized for position control of each joint of the manipulator independently. Each joints' controller is based on single input and single output (SISO) system. There in the first case, the performance of the controller on each joint has been evaluated for open-chain system with floating base and locked telescopic joints. For the second case, the performance of the controller in close chain, when the telescopic joints are released, has been investigated. After reconfiguration the value of joints will be considered for both cases and some numerical results will be reported.