DOGZILLA is a visual AI robot dog with 12 degrees of freedom, primarily built from 6 servos, aluminum brackets, and a camera. It can flexibly perform a range of bionic actions, achieve omni-directional movement, and exert six-dimensional attitude control. DOGZILLA is equipped with an IMU and servo-angle sensors to provide real-time feedback on its posture and joint angles. Using inverse kinematics, it achieves a variety of motion gaits. Raspberry Pi serves as its main controller, with optional lidar and voice modules. Through Python programming and the ROS2 framework, it enables AI visual recognition, lidar-based mapping and navigation, and voice control.
In the new upgrade, DOGZILLA introduces AI large-language-model capabilities, giving the robot dog enhanced understanding and interactivity. Based on the AI large model, DOGZILLA can perform complex semantic speech recognition and natural language dialogue, intelligently analyze visual scenes, plan paths, comprehend task instructions, and support more sophisticated and anthropomorphic behavioral decisions. This upgrade significantly expands the application depth and scalability of robot dogs in education, scientific research, AI experiments, service robotics, and related fields, helping users embark on a new journey of embodied intelligence.
Feature
- Can walk and twist like a real dog.
- Equipped with 12PCS high-precision servo motors, a safe and non-toxic aluminum alloy body, and a wide-angle camera. S2 has added lidar and voice interaction modules.
- Using Raspberry Pi as the controller, we have upgraded the ROS2 system to support Python programming and RVIZ simulation.
- Support multiple remote control methods such as APP, handle, web pages, computer keyboards, and APP mapping navigation.
- S1/S2 is easily completed based on ROS2 and OpenCV, with functions such as label recognition, face detection, target tracking, and visual line patrol.
- S2 comes with a lidar and intelligent voice module, which can achieve functions such as mapping navigation, lidar avoiding and following, and voice control.
- Based on the AI larger model and AI voice recognition module, DOGZILLA can understands commands, and achieve related functions such as voice dialogue, scene perception, embodied intelligence, etc.