Sahasranshu Technologies / PS 01
Space Domain Awareness
For space asset protection and orbital sustainability
Build an AI/ML-powered Space Domain Awareness platform that tracks orbital objects with LiDAR and predicts collision risks with quantified uncertainty.
Space Domain Awareness
The objective
The challenge.
The rapid proliferation of satellites, space debris and unidentified orbital objects in Low Earth Orbit (LEO) is a growing challenge to the safety, security and sustainability of space operations. Existing Space Domain Awareness systems struggle to track smaller objects, maintain accurate orbital trajectories, detect anomalous movements and deliver timely collision-risk assessments. Build an indigenous, AI/ML-powered Space Domain Awareness platform that leverages ground-based and space-based LiDAR technologies, integrated with optical and radar sensors, to enable accurate detection, tracking, characterization and predictive analysis of orbital objects.
Satellite operators and national space agencies who need real-time situational awareness, visual analytics and decision support; space-traffic coordinators tracking assets in increasingly crowded orbits.
A validated technology demonstrator that improves orbital tracking accuracy, reduces false collision alerts and enables earlier identification of potentially hazardous space events — establishing an indigenous, AI-enabled Space Domain Awareness capability that strengthens India’s space asset protection, space traffic coordination and long-term orbital sustainability.
Product capabilities
What to build.
Required features
- LiDAR-based object tracking
Leverage high-precision laser-ranging measurements to improve the positional accuracy of satellites and orbital debris.
- AI/ML-powered sensor fusion
Integrate LiDAR, radar, optical observations and orbital catalogues into a unified, continuously updated space-object tracking picture.
- Predictive orbital intelligence
Develop physics-informed machine learning models to estimate orbital trajectories, predict close approaches and assess collision risks with quantified uncertainty.
- Anomaly detection
Identify unusual orbital manoeuvres, unexpected trajectory deviations and potentially hazardous proximity events.
- Autonomous space surveillance
Enable intelligent sensor scheduling, observation prioritization, automated object association and explainable risk alerts.
- Indigenous SDA command platform
Build a scalable, secure, on-premises platform providing real-time situational awareness, visual analytics and decision support for satellite operators and national space agencies.
Optional ideas
- Catalogue federation
Correlate additional orbital catalogues and space-environment feeds to enrich the unified tracking picture.
- Constellation-scale monitoring
Extend the tracking and prediction pipeline from individual objects to full constellations.
- Deeper explainability
Richer explanations for risk alerts, showing which sensor observations drove each prediction.
Typical workflow
LiDAR, radar and optical observations flow in → sensor fusion builds a unified, continuously updated space-object picture → physics-informed models estimate trajectories, predict close approaches and quantify collision risk → anomalies and hazardous proximity events are flagged → the command platform delivers real-time situational awareness and decision support to operators.
Demo & deliverables
Show it working.
- SDA prototype
An end-to-end demonstrator covering detection, tracking, characterization, sensor fusion and predictive analysis of orbital objects.
- Predictive intelligence report
Orbital trajectory estimates, close-approach predictions and collision-risk assessments with quantified uncertainty.
- Command platform demo
The on-premises platform showing real-time situational awareness, visual analytics and decision support for operators.
- Evaluation report
Tracking accuracy, false-alert reduction, anomaly-detection performance and uncertainty quantification from your test set.
Success criteria
What success looks like.
Improved positional accuracy for satellites and orbital debris through LiDAR-based ranging.
Reliable orbital trajectory estimates, close-approach predictions and collision risks with quantified uncertainty.
Unusual manoeuvres, trajectory deviations and hazardous proximity events identified early.
Fewer false collision alerts compared with the baseline.
A scalable, secure, on-premises command platform delivering real-time situational awareness.
Scope & constraints
Build within the brief.
Detection, tracking, characterization and predictive analysis of LEO orbital objects using LiDAR, radar, optical sensors and AI/ML — with an on-premises command platform for operators.
Physical sensor hardware development is not required — teams demonstrate the platform on provided formats, simulated or publicly available observation data.
The partner supplies no live sensor streams; generate realistic LiDAR, radar and optical observations to develop and demonstrate the system.
Technical & safety requirements
LiDAR, radar, optical observations and orbital catalogues must fuse into one continuously updated tracking picture.
Predictive models must be grounded in orbital physics and report uncertainty, not just point estimates.
The command platform must be scalable, secure and run on-premises for satellite operators and national space agencies.
Sensor scheduling, observation prioritization and object association should be automated with explainable risk alerts.
Resources & help
Useful before you start.
Orbital catalogue data for object association and the unified tracking picture.
Laser-ranging measurements to improve positional accuracy of satellites and debris.
Radar and optical observations to fuse with LiDAR into the tracking picture.
Clarify data formats, sensor models and platform requirements with the challenge desk.
