An end-to-end autonomous driving approach that separates semantic and motion learning into parallel decoders to mitigate negative transfer.

A scene-wide motion forecasting approach that transforms local agent-centric embeddings via a latent context module.

This paper subsumes our IV 2018, ITSC 2018, and RSS 2020 workshop papers.

A discussion including the role of smart mobility solutions in sustainable smart districts.

We combine LTV-MPC with constrained RL to generate safe reference trajectories for motion planning.

My dissertation formulates motion planning under uncertainty and solves it using nonlinear MPC and Monte Carlo tree search.

A sharpness-continuous path optimization and sparsification approach, ensuring smooth motion under curvature and actuator constraints.

We design and experimentally validate an LTV-MPC for lateral control in truck-trailer systems, achieving precise and stable reversing maneuvers.

A game-theoretic motion planning algorithm that models vehicle interactions as a Stackelberg game, solved efficiently through single-level reformulation.

This paper is subsumed by our journal paper.

This paper is subsumed by our journal paper.

This paper is subsumed by our journal paper.

A comparative analysis of trajectory similarity measures for prediction evaluation.

An overview of our autonomous driving platform that won first place in the technical category and second place overall at the GCDC 2016.

A hierarchical taxonomy for rating cooperative automated driving systems into information-based and maneuver-based levels.

Building on the takeaway of our IV 2016 paper, it presents a system architecture that integrates performance assessment at every layer.

A review of autonomous vehicle system architectures, highlighting how self-aware software modules enhance robustness and reliability.

An evaluation of ROS for autonomous driving, analyzing its architecture and performance under high data load.

A refinement of my master's thesis with improved formalization, an alternative solver, and a different initialization approach.

A motion planning framework that partitions trajectory space into multiple solution classes, enabling local optimization to find globally optimal solutions.

Cooperative Adaptive Cruise Control implementation of Team Mekar's GCDC 2011 vehicle that uses RTK GPS, IEEE 802.11p, and electric motor brake actuation for automated longitudinal control.