The obstacle avoidance system incorporates TensorFlow SSD MobileNet v2 model on a Raspberry Pi 5. It identifies objects including people, cars, lamp poles, drains, bicycles, and motorcycles and informs the users of the obstacles on their path. The Raspberry Pi Camera Module V3 captures images and the trained model is used to process these images and generate detection results. The dataset that was used in training the model was split into training, validation, and test sections in order to achieve high levels of accuracy when identifying obstacles. Furthermore, the RTN is possible to navigate to 5 daily needs destinations: toilet, mosque, home, bus stop, restaurant. The RTN tracks navigation paths in the real world using a camera and gives users with real-time navigational instructions via acoustic speakers, allowing users to safely go from one location to another. The RTN features a voice assistant with TTS and STT capabilities that users can interact with by speaking directly to the MEMS microphone within three seconds through voice commands. The ability of searching for the nearest relevant destinations coordinates and paths using keywords with Openroute Service’s API without specific the exact name of the destinations with a search radius ranging from 0km to 1km, it eliminates the need for users to remember the location’s name, thus making the navigation process even more intuitive and efficient, and enhancing user’s navigation experiences. The Hotlink 4G LTE SIM card makes it possible that ObstaSense is always connected to the API servers for data monitoring and communication in real-time, accomplished by the Lilygo T-A7670G ESP32 MCU.