Max Robotics — Global Service Robot Directory + US Certification

About

Sourav Das is a researcher specializing in human trajectory prediction, a critical challenge at the intersection of computer vision, machine learning, and robotics. His work focuses on developing intelligent models capable of forecasting pedestrian movement across varying time horizons, with direct applications in autonomous vehicles, socially aware robots, and advanced surveillance systems. His most influential contribution, "Goal-driven Self-Attentive Recurrent Networks for Trajectory Prediction" (2022, 61 citations), introduced a framework that integrates past motion, environmental context, and destination estimation to address the inherently multi-modal nature of human movement. This work has become a notable reference in the trajectory forecasting community. Building on this foundation, Das has more recently explored knowledge distillation as a technique for bridging short-term and long-term prediction, tackling the increasing uncertainty that emerges over extended time horizons. His 2025 work, "KD-Mamba," further advances this direction by integrating selective state space models with knowledge distillation, reflecting his engagement with cutting-edge sequence modeling architectures. With a growing publication record and rising citation impact, Das represents an emerging voice in trajectory forecasting research, consistently pushing toward more robust, scalable, and temporally aware prediction systems.