Nishanth Kumar

Papers

Total Citations

H-Index

About

Nishanth Kumar is an emerging researcher at the intersection of robot planning, abstraction learning, and neuro-symbolic artificial intelligence. His work addresses one of the fundamental challenges in autonomous robotics: enabling robots to plan efficiently across continuous, complex state and action spaces. Kumar's most recognized contribution, "Predicate Invention for Bilevel Planning" (2023, 15 citations), introduces a framework where high-level abstract plans guide low-level planning, significantly reducing the computational burden of planning in otherwise intractable environments. Building on this foundation, his 2024 work "VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planning" advances the field by proposing a first-order abstraction language that marries the interpretability of symbolic reasoning with the perceptual power of neural networks. This approach allows robots to form task-specific world models that filter irrelevant sensorimotor complexity while retaining task-essential structure — a critical step toward broadly intelligent agents. With growing citation counts and a research agenda that bridges perception, abstraction, and planning, Kumar represents a promising voice in next-generation robot learning and AI systems design.