Giovanni Sutanto

Papers

Total Citations

H-Index

About

Giovanni Sutanto is a robotics researcher whose work sits at the intersection of differentiable physics, robot learning, and constrained motion planning. His primary research areas include differentiable robot dynamics, learning-based motion planning on manifolds, and the integration of physical constraints into neural network architectures. Sutanto’s most notable contribution is his work on the differentiable Newton-Euler Algorithm (RNEA), which encodes physical constraints directly into computational graphs, enabling robots to learn their own dynamics parameters from data using modern auto-differentiation toolboxes—a paper that has garnered 11 citations. He has also made significant strides in learning constraint manifolds for sequential motion planning, exploring methods such as Variational Autoencoders (VAE) and introducing the Equality Constraint Manifold Neural Network (ECoMaNN) to learn representations of constraints from demonstrations. His 2022 work on differentiable and learnable robot models further advances the merging of rigid-body simulations with differentiable machine learning. Sutanto’s research is foundational for building robots that can autonomously adapt their motion and dynamics in complex, real-world environments.