Papers
23
Total Citations
458
H-Index
11
About
Boris Ivanovic is a prominent researcher at the intersection of autonomous systems, multi-agent behavior prediction, and human-robot interaction. His work centers on enabling robots and self-driving vehicles to reason about the complex, uncertain futures of surrounding agents — a capability fundamental to safe navigation in real-world environments. Ivanovic is perhaps best known for the Trajectron series of models, beginning with "The Trajectron" (2019) and culminating in "Trajectron++" (2020), which has garnered over 150 citations across its variants. These frameworks introduced probabilistic, graph-structured generative models capable of forecasting the trajectories of heterogeneous agents using diverse data sources, setting a new standard in the field. His subsequent work, including BITS (2023) and MATS (2020), extends this vision into realistic traffic simulation and planning-compatible prediction representations. A recurring theme across his research is uncertainty — from propagating state uncertainty through forecasting pipelines to incorporating class ambiguity into predictions. His 2022 work on conformal prediction further demonstrates his commitment to rigorous safety assurance. Through consistently high-impact publications, Ivanovic has helped bridge the gap between theoretical behavior modeling and practical autonomous driving deployment, making him a leading voice in the robotics and autonomy research community.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3BITS: Bi-level Imitation for Traffic Simulation54 citations · 2023
- 4Heterogeneous-Agent Trajectory Forecasting Incorporating Class Uncertainty38 citations · 2022
- 5
- 6
- 7Sample-Efficient Safety Assurances Using Conformal Prediction25 citations · 2022
- 8Injecting Planning-Awareness into Prediction and Detection Evaluation21 citations · 2022
- 9Propagating State Uncertainty Through Trajectory Forecasting20 citations · 2022
- 10