MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202631061048 A) filed by National Institute Of Technology, Aizawl, Mizoram, on May 13, for 'system and method for generating a behaviour conditioned interactive synthetic trajectory dataset for indoor scenarios.'
Inventor(s) include Malviya Vaibhav; Tripathi Vishnu Dev; Malviya Abhinav; and Kala Rahul.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "The present invention generally relates to a system (100) for generating a behaviour-conditioned interactive synthetic trajectory dataset for indoor environments. One or more sensing nodes (102) acquire three-dimensional LiDAR point cloud frames that are temporally stored by a frame buffering unit (104). A motion extraction unit (106), refined by a noise suppression unit (108), isolates moving human agents in a calibrated world coordinate system. An agent segmentation unit (110) delineates individual agents and assigns instance labels, and a trajectory construction unit (112) generates continuous trajectories. A behaviour parameter extraction unit (114) derives behaviour-specific separation and waiting parameters from interacting trajectories. A microscopic motion synthesis unit (116), combined with a behaviour fusion unit (118), generates behaviour-consistent motion updates. A navigation unit (120) ensures obstacle-free movement using waypoint-based paths. A dataset generation unit (122) executes repeated simulations and records agent states and behaviour labels to produce a synthetic indoor crowd trajectory dataset."
Disclaimer: Curated by HT Syndication.
