The rapid aging of populations in developed countries, including Germany, presents significant challenges, particularly concerning dependency among older adults. This paper addresses the critical need for assistive technology to aid older individuals in performing simple daily tasks, focusing on developing a semi-automatic robotic system using the Unitree Go1 platform. The research aims to enhance the quality of life for older adults by enabling them to interact with objects in their environment more independently. By utilizing a depth camera, the system estimates object geometry and location, planning the robot's body and arm movements for effective object manipulation. The study integrates various subsystems, leveraging existing libraries and community-developed packages to implement a grabbing strategy. Multiple approaches are developed and assessed based on behavior and accuracy, with the best-performing method achieving an 84% success rate in varied testing conditions. This paper provides a detailed review of research and implementation, potentially accelerating future research and product development in assistive robotics.