WO2024052928 - SYSTEM FOR VISION-BASED SELF-DECISIVE PLANETARY HAZARD FREE LANDING OF A SPACE VEHICLE
National phase entry is expected:
Publication Number
WO/2024/052928
Publication Date
14.03.2024
International Application No.
PCT/IN2023/050811
International Filing Date
28.08.2023
Title **
[English]
SYSTEM FOR VISION-BASED SELF-DECISIVE PLANETARY HAZARD FREE LANDING OF A SPACE VEHICLE
[French]
SYSTÈME D'ATTERRISSAGE PLANÉTAIRE SANS DANGER, VISUEL ET AUTODÉTERMINANT D'UN VÉHICULE SPATIAL
Applicants **
PATIL, Dipti
Flat 303, Ankoorshree CHS, Happy Colony, Lane 3, Kothrud
Pune 411038, IN
BORSE, Janhavi
44- Pasay, Sahadev Nagar, Gangapur Road,
Nashik 422013, IN
KUMAR, Vinod
Flat No. 2D, COSMOS Klassik Landmark, Kasavanahalli
Banglore 560035, IN
Inventors
PATIL, Dipti
Flat 303, Ankoorshree CHS, Happy Colony, Lane 3, Kothrud
Pune 411038, IN
BORSE, Janhavi
44- Pasay, Sahadev Nagar, Gangapur Road,
Nashik 422013, IN
KUMAR, Vinod
Flat No. 2D, COSMOS Klassik Landmark, Kasavanahalli
Banglore 560035, IN
Priority Data
202221051791
10.09.2022
IN
Application details
| Total Number of Claims/PCT | * |
| Number of Independent Claims | * |
| Number of Priorities | * |
| Number of Multi-Dependent Claims | * |
| Number of Drawings | * |
| Pages for Publication | * |
| Number of Pages with Drawings | * |
| Pages of Specification | * |
| * | |
| * | |
International Searching Authority |
IP India
* |
| Applicant's Legal Status |
Natural Person
* |
| * | |
| * | |
| * | |
| Entry into National Phase under |
Chapter I
* |
| Translation |
|
Recalculate
* The data is based on automatic recognition. Please verify and amend if necessary.
** IP-Coster compiles data from publicly available sources. If this data includes your personal information, you can contact us to request its removal.
Quotation for National Phase entry
| Country | Stages | Total | |
|---|---|---|---|
| China | Filing | 963 | |
| EPO | Filing, Examination | 4861 | |
| Japan | Filing | 588 | |
| South Korea | Filing | 574 | |
| USA | Filing, Examination | 2710 |
Total: 9696 USD
The term for entry into the National Phase has expired. This quotation is for informational purposes only
Abstract[English]
The present invention relates to a system for vision based self-decisive planetary hazard free landing of a space vehicle. The invention discloses the method of finding correlations between descent images captured by cameras and the system state parameters to predict the thrust control command for spacecraft descent guidance and navigation. For hazard-free landing, a hazard detection system (105) is activated, wherein the terrain under the current field of view is classified into craters, boulders, and plane surface, which is then used by thrust prediction utility (104) to take retargeting decisions. The module (103) discloses the method of generating a synthetic dataset of planetary images, and a descent trajectory state parameters through an agent-based image generative platform for an unknown planetary environment. Multi-variate deep learning models are used to predict the control actions in the form of thrust command (106) by combining the results of models (103) and (105).[French]
La présente invention concerne un système d’atterrissage planétaire sans danger, visuel et autodéterminant d'un véhicule spatial. L'invention divulgue le procédé consistant à rechercher des corrélations entre des images de descente capturées par des caméras et les paramètres d'état de système pour prédire l’instruction de commande de poussée permettant un guidage et une navigation de descente d'engin spatial. Pour un atterrissage sans danger, un système de détection de danger (105) est activé, le terrain sous le champ de vision actuel étant classé en cratères, rochers et surface plane, qui est ensuite utilisé par l'utilitaire de prédiction de poussée (104) pour prendre des décisions de reciblage. Le module (103) divulgue le procédé de génération d'un ensemble de données synthétiques d'images planétaires, et un paramètre d'état de trajectoire de descente par l'intermédiaire d'une plateforme générative d'image utilisant un agent destiné à un environnement planétaire inconnu. Des modèles d'apprentissage profond à variables multiples sont utilisés pour prédire les actions de commande sous la forme d'une instruction de poussée (106) par combinaison des résultats de modèles (103) et (105).