- Research and development
- Inspection and maintenance
- Patrol and detection
- Last-mile delivery
- Exploration and rescue
- The patented design is based on rigorously tested self-developed technology so that Aliengo is calibrated to derive maximum performance for the longest possible time, while it’s integrated body is made of polymer plastic, carbon fibre and aviation-grade aluminium.
- Unitree Robotics has developed a lithium battery pack with an incredible 4.5 hour operating time, allied with an intelligent built-in air cooling system that automatically senses the temperature and cools accordingly.
- Advanced dynamic balancing algorithm enabling swift regain of balance following impact or falls.
- Force control technology grants compound control of the robot dog’s joints, realising full control of triple-axis posture and position for multi-terrain adaptability. Aliengo can run with equally strong stability on rugged gravel roads or grass.
- 12 high-performance servo motors.
- Maximum walking speed in excess of 1.5 m/s
- A highly agile quadruped robot capable of numerous special movements:
- high speed running
- running backwards
- lateral movement
- in-situ turning
- creeping forward
- obstacle crossing
- obstacle avoidance
- climbing in place following a fall
- ascent and descent of slopes and steps
- Robust, shockproof fuselage stands up to significant impacts.
Rich open system
- Sofware control interface divided into high level and bottom level
- Control interface supports C / C++, ROS and more
- Allows reading of numerous types of sensor data
- Allows control of single motor or single leg movement
Rich external interface
- Developer version with onboard PC and corresponding open interface
- External interface: 3 x USB, 2 x HDMI, 2 x Ethernet, 1 x 485 port
- Integrate a range of interactive systems for numerous applications: vision camera, LiDAR, robotic arm, GPS
Multi eye intelligent depth camera
- Global shutter and wide field of view
- Minimum sensing depth – 0.11 m
- 1280 x 720 depth resolution
Visual odometer camera
- Highly optimised V-SLAM
- Closed-loop offset <1%
- <6ms delay between gesture motion and motion reflection
- Fisheye lens imager combined with near hemispherical FoV (165° +/- 5°) enables fast-tracking with fast movement
Depth vision – real-time 3D map creation and navigation planning
3D environment construction – while in motion Aliengo uses the cameras to obtain colour and depth information of the environment, then reconstructs the 3D spatial information of objects with the help of a specific vision algorithm.
Probability map – octomaps / probability maps are created bu the cameras as the robot moves, detecting surroundings and providing obstacle data.
Dynamic obstacle perception – when the robot encounters a dynamic obstacle it refreshes that map data within a particular range, discarding the ‘moving artefact’ left by the dynamic obstacle on the map.
Global positioning – global and local real-time positioning functions are available during the process of map creation. The map will follow the camera’s perspective in real-time, supporting real-time zoom, movement and rotation.
Loop detection – Aliengo maintains high loop-back accuracy in a wide range of fields, high positioning accuracy within a certain range, maintaining stability with a certain oscillation amplitude, with drift or loss.
Human posture recognition tracking and face recognition
Body posture recognition – Aliengo’s deep learning module enables the colour camera to identify specific human postures and conduct human-machine interaction, the robot can then make movements that correspond to different body postures.
Human skeleton perception – the robot can analyse and calculate the two-dimensional skeleton information of the human body according to colour information from perspective, and analyse and calculate the three-dimensional skeleton information and motion of a specific character using depths of field.
Target person tracking – when multiple pedestrians are present, Aliengo can be instructed to lock on to an individual by a particular posture (eg raising the left hand), the robot will then follow the movement of the target,