DOGZILLA is a visual AI robot dog with 12 degrees of freedom, mainly composed of 6 servos, aluminum alloy brackets, and a camera. It can flexibly complete a series of bionic actions and achieve omni-directional movement and six dimensional attitude control. DOGZILLA is equipped with IMU and servo angle sensors, which can provide real-time feedback on its own posture and joint angles. With inverse kinematics algorithms, achieves various motion gaits. Raspberry Pi as its main controller, with additional configurations such as lidar and voice module. Through Python programming, based on the ROS2 system, many functions such as AI visual recognition, lidar mapping navigation, and voice control can be achieved.
With the New upgrade, DOGZILLA has introduced AI large language model capabilities, giving the robot dog stronger understanding and interactivity. Based on the AI large model, DOGZILLA can not only perform complex semantic speech recognition and natural language dialogue, but also intelligently analyze visual scenes, plan paths, understand task instructions, and support more intelligent and anthropomorphic behavioral decisions. This upgrade has greatly improved the application depth and scalability of robot dogs in education and scientific research, artificial intelligence experiments, service robots and other fields, helping users embark on a new journey of exploration of embodied intelligence.