AstrAlive
About
About AstrAlive
AstrAlive is a Beijing company developing data infrastructure for embodied AI, with a focus on first-person recordings of human manipulation. Its EgoKit capture system combines visual, hand-pose and tactile signals. The HBR Engine processes those recordings through hand-pose estimation, multimodal fusion and automated annotation, turning interactions into reusable training assets. The company offers standardized datasets and project services spanning collection design, equipment deployment and delivery to model developers and robot manufacturers.
Its engineering work addresses occlusion, image distortion, rapid movement and difficult lighting during hand–object interactions. AstrAlive reports project-level cooperation with robotics and embodied-model companies. A partnership with LUSTER subsidiary Yuanke Vision combines its processing software with first-person capture, multi-view vision and optical motion-capture hardware. Its current commercial focus is building consistent data quality and repeatable delivery across customer projects.
Funding
- 2026-10-22 - Seed and seed-plus - AstrAlive completes seed and seed-plus financing for embodied-AI data
AstrAlive announced tens of millions of yuan in combined seed and seed-plus financing to expand its Physical AI data infrastructure. - source: news.pedaily.cn
- 2026-03-27 - First financing - AstrAlive completes first financing led by Shuimu Capital
AstrAlive announced first financing in the tens of millions of yuan for its first-person embodied-data technology. - source: media.qimingpian.cn
Investors
- First Capital Investment - Participant - VC - 1 funding event - first seen 2026-10-22
- LUSTER - Lead - Unknown - 1 funding event - first seen 2026-10-22
- Quanshi Capital - Participant - VC - 1 funding event - first seen 2026-03-27
- Shuimu Capital - Lead - VC - 1 funding event - first seen 2026-03-27
- Shunxi Fund - Lead - VC - 1 funding event - first seen 2026-10-22
- Yaozhuo Capital - Participant - VC - 1 funding event - first seen 2026-03-27
- Zhongguancun Capital - Participant - VC - 1 funding event - first seen 2026-10-22
Team
- 宋知珩 - Founder & CEO
Timeline
Events
- 2026-10-22 - Funding - AstrAlive completes seed and seed-plus financing for embodied-AI data
Shunxi Fund led the seed round; LUSTER led seed-plus, joined by Zhongguancun Capital and First Capital Investment. Capital supports algorithms and products, real-world multimodal data infrastructure, customer delivery and commercialization. LUSTER subsidiary Yuanke Vision will combine capture and optical motion-capture hardware with AstrAlive’s automated data processing.
Source - 2026-03-27 - Funding - AstrAlive completes first financing led by Shuimu Capital
Shuimu Capital led the round; Yaozhuo Capital, Quanshi Capital and industry angels also participated. Its capture hardware and data engine combine visual, tactile and pose signals to turn human interactions into training data.
Source
Relations
Relationships
- Luster
Partner - AstrAlive and LUSTER subsidiary Yuanke Vision are combining hand-interaction data processing with first-person, multi-view and optical motion-capture hardware.
Context
Comparable profiles
- Lightwheel
stack adjacency - Robotics Data Infrastructure Layer
- Synthium
stack adjacency - Robotics Data Infrastructure Layer
- Xspark AI
stack adjacency - Robotics Data Infrastructure Layer
- Mecka AI
stack adjacency - Robotics Data Infrastructure Layer
- shotwell.ai
stack adjacency - Robotics Data Infrastructure Layer
- Encord
stack adjacency - Robotics Data Infrastructure Layer
- 6thSense
stack adjacency - Robotics Data Infrastructure Layer
- Cortex AI
stack adjacency - Robotics Data Infrastructure Layer
- Physical Turing
stack adjacency - Robotics Data Infrastructure Layer
- Valgo
stack adjacency - Robotics Data Infrastructure Layer
- Sensei
stack adjacency - Robotics Data Infrastructure Layer
- Sureform
stack adjacency - Robotics Data Infrastructure Layer
Shared workflows
- LabyrinthAI
workflow adjacency - Multimodal Robotics Data Labeling · Robot Learning Data Collection
- Yitai Embodied
workflow adjacency - Multimodal Robotics Data Labeling · Robot Learning Data Collection
- Mind Robotics
workflow adjacency - Robot Learning Data Collection
- SenseGlove
workflow adjacency - Robot Learning Data Collection
- Vision Lab
workflow adjacency - Robot Learning Data Collection
- MANUS
workflow adjacency - Robot Learning Data Collection
- Trossen Robotics
workflow adjacency - Robot Learning Data Collection
- Rokoko
workflow adjacency - Robot Learning Data Collection
- Noitom
workflow adjacency - Robot Learning Data Collection
- StretchSense
workflow adjacency - Robot Learning Data Collection
- XDOF
workflow adjacency - Robot Learning Data Collection
- Maniformer
workflow adjacency - Robot Learning Data Collection
Context tags
- Robotics data infrastructure layer - Stack Layer - Capture, HBR Engine processing and data-delivery infrastructure: https://xy-ai.tech/blog/ . Data provider, not robot/model developer.
- Robot learning data collection - Workflow - EgoKit human-manipulation capture for robot learning: https://xy-ai.tech/blog/ .
- Multimodal robotics data labeling - Workflow - Multimodal automated annotation and quality assessment: https://xy-ai.tech/blog/ .