Rajas Bansal

About

Rajas Bansal is a researcher advancing the frontier of robotic manipulation and common-sense reasoning. His work centers on enabling robots to generalize tool use in dynamic, real-world environments—a critical step toward autonomous assistants in factories and homes. In his highly cited paper "TOOLTANGO: Common Sense Generalization in Predicting Sequential Tool Interactions for Robot Plan Synthesis" (2022), Bansal tackles the challenge of teaching machines when and how to employ objects as tools, such as using a tray to carry items. This research bridges the gap between low-level motor control and high-level planning, demonstrating how robots can compose sequences of tool interactions without explicit programming. Though early in his career, his contributions have already garnered attention (4 citations), signaling a growing impact in the robotics community. Bansal’s work is notable for its focus on common-sense generalization—a key bottleneck in embodied AI—and his findings lay the groundwork for more adaptable, intelligent robotic systems. For students and researchers, his approach offers a compelling model of how to integrate cognitive principles with practical robot learning.

Research Focus

Key Achievements

Top Papers

1. 1
   TOOLTANGO: Common sense Generalization in Predicting Sequential Tool Interactions for Robot Plan Synthesis

   4 citations · 2022

Key Collaborators

Contact & Links