Early detection of tumors remains a critical challenge in oncology, necessitating the development of innovative and efficient detection techniques. This study proposes a novel approach for tumor targeting employing a swarm of nanorobots (NS) and their light-driven aggregation capabilities. By exploiting the interaction behavior of NS aggregation under light exposure, we aimed to improve tumor detection capabilities. The effectiveness of this approach was assessed through NS dispersion and search efficiency under various light conditions, including both periodic and aperiodic sources. The complex tumor microenvironment, characterized by a dense capillary network, leads to the formation of biological gradient fields (BGFs) within Manhattan-geometry vasculature (MGV). This research thoroughly investigates the behavior of NS within these specific environments, mimicking the navigational constraints imposed by MGV. Our findings validate the feasibility of utilizing light-driven NS aggregation for precise and efficient tumor targeting. The results establish a foundation for innovative tumor detection methodologies, emphasizing the significant potential of NS in clinical applications, especially within navigating MGV and responding to BGFs.