Embodied Intelligence Software System |
Product Name | Module | Function | Description |
Dashu Embodied Intelligence Agent Application System | Motion & Dance | Motion and dance library | Includes a built-in library of performance motions, humanoid gestures and dances, such as handshakes, waving, greetings,
high-fives, martial arts-style actions and themed dance routines. |
Voice Interaction | Voice interaction module | Supports interactive dialogue through the app or microphone wake-up. The robot can perform multi-turn conversation,
voice-controlled actions, dances, navigation guidance, basic movement commands and knowledge-base Q&A. |
Version 1.0: Supports Chinese and English language configuration and basic voice interaction.
Version 2.0: Adds end-to-end voice interaction, faster response, Chinese/English switching, selected dialect support, singing and
more natural emotional voice expression. |
Supports configurable voice agents for different reception, guidance and tour scenarios, including greeting scripts and background
information. |
Supports voice wake-up and interruption during conversations. |
Autonomous Navigation
(No Remote Controller Required) | Hardware configuration | Bluetooth speaker, wireless microphone set, and visual module with camera accessories can be configured according to the
application scenario. |
Quick mapping | The robot can quickly scan the surrounding environment by remote control and build a high-precision map for navigation and
obstacle avoidance. |
Autonomous navigation | Uses path-planning algorithms to create efficient tour routes based on the mapped environment. |
Custom tasks | Supports customized guide and reception workflows, including navigation points, fixed-point explanations, actions, photo/video
capture, pause, resume, stop and loop operations. |
Third-party interface integration | Supports integration with external systems such as display screens, lighting systems and custom message interfaces within task
workflows. |
Autonomous obstacle avoidance | Uses multi-sensor fusion to identify obstacles and avoid them flexibly for safer operation. |
Web admin panel | Provides a web-based management panel for remote configuration, monitoring, scenario agent setup and knowledge-base upload. |
App operation | Provides a simple app interface for task execution, voice interaction, remote mapping, point setup and map switching. It also
supports remote movement control and viewing the robot's visual feed. |
Visual recognition and analysis | Supports face and pedestrian recognition for greeting scenarios, and can use visual AI models for scene recognition and analysis. |
AI model configuration | Supports OpenAI-compatible model interfaces and built-in speech and language models for more natural voice interaction. |
Group control | Supports unified control of multiple robots for synchronized actions, dances and voice broadcasts. |
Precise positioning and navigation | Uses visual understanding to obtain spatial information and improve positioning and navigation accuracy in dynamic environments. |
Operations & After-Sales Support | Software support
(Optional service period) | Annual software support | Includes software issue fixes, software updates when available, and cloud service support. |
LLM Tokens | LLM token package
(Optional service period) | Token service | Optional token packages are available for AI model usage and can be expanded as needed. |
On-Site Implementation | Implementation service | | Includes on-site implementation by one technical engineer for up to four days, covering mapping, point setup, workflow
configuration, system training, navigation debugging and voice interaction debugging. |
Value-Added Services for Humanoid Embodied Intelligence System |
Product Name | Module | Function | Description |
Value-Added Service 1
(Optional) | Mini Program
(Visitor Side) | Mini program | Enables visitors to interact with the robot through Q&A, motions and dance commands. Suitable for public venues such as museums,
hospitals and science centers. |
Value-Added Service 2
(Optional) | Customization | Private software deployment | Supports private deployment authorization, voice service deployment, backend service deployment and on-site implementation.
Available as an optional extension to the basic package. |
Value-Added Service 3
(Optional) | PPT Presentation
(Optional) | PPT presentation capability | Supports robot-led presentations. The robot can interact with the audience before class, present content with synchronized PPT
page turning, answer questions during the session and then resume the presentation workflow. |
Value-Added Service 4
(Optional) | Multimodal Function
(Optional) | Visual understanding and vision-based interaction | Supports multimodal AI models and visual agents for natural voice interaction based on visual understanding. The robot can
recognize objects, understand scenes, identify people, assess tasks and respond by voice.
In practical use, it can support roles such as safety supervision or reception. It can identify abnormal situations, issue voice
reminders and follow preset response steps.
The system can also import images, documents and customer information in advance, enabling short-term visual memory and contextual
reception, such as recognizing a visitor and providing a personalized greeting and introduction. |
