Harshayu Girase

Papers

Total Citations

H-Index

About

Harshayu Girase is a researcher whose work lies at the critical intersection of computer vision and autonomous driving, with a primary focus on trajectory prediction. His most influential contribution, "It Is Not the Journey But the Destination: Endpoint Conditioned Trajectory Prediction" (2020), which has garnered 43 citations, introduces a paradigm shift in how autonomous systems anticipate the future motion of traffic actors. Rather than optimizing entire paths, Girase's approach cleverly conditions predictions on the likely final destination of a vehicle or pedestrian, dramatically improving long-term forecasting accuracy. Building on this foundation, his subsequent work, "Physically Feasible Vehicle Trajectory Prediction" (2021), addresses a crucial gap in the field by ensuring that predicted trajectories respect the fundamental laws of physics—a property often overlooked by models that merely optimize standard error metrics. By explicitly incorporating constraints like acceleration limits and kinematic feasibility, Girase bridges the gap between data-driven prediction and real-world applicability. His research is essential reading for anyone working on safe, reliable autonomous navigation systems, demonstrating that the most effective predictions are those that understand both where an agent is going and how it can physically get there.