About

David Sierra-Gonzalez is a leading researcher in autonomous vehicle perception, specializing in sensor fusion and real-time environmental representation. His work centers on developing robust methods for ground plane estimation and semantic grid prediction—critical components for enabling safe navigation in robotics and self-driving cars. His most influential contribution, **GndNet** (2020, 88 citations), introduced a fast deep-learning approach for simultaneous ground plane estimation and point cloud segmentation, directly addressing a fundamental bottleneck in 3D object detection, occupancy mapping, and localization. Building on this, he pioneered multimodal fusion techniques: **TransFuseGrid** (2022) was among the first to combine LiDAR and RGB data via transformers for semantic grid prediction, overcoming the limitations of vision-only systems. His latest work, **LAPTNet-FPN** (2023), advances real-time multi-scale LiDAR-aided projective transforms, enabling efficient semantic scene understanding for navigation and tracking. By bridging the gap between raw sensor data and actionable spatial representations, Sierra-Gonzalez’s research directly impacts the reliability of autonomous systems in dynamic environments. His contributions are shaping the next generation of perception stacks for intelligent vehicles.

Research Focus

Key Achievements

3
H-Index
3
Papers
102
Total Citations
34
Avg Citations/Paper
🏆 Most Cited Paper
GndNet: Fast Ground Plane Estimation and Point Cloud Segmentation for Autonomous Vehicles
88 citations · 2020
📈 Most Prolific Year: 2020 (1 Papers)
🤝 Key Collaborators: 7
🏛 Institutions: Centre Inria de l'Université Grenoble Alpes, Université Grenoble Alpes, Institut national de recherche en sciences et technologies du numérique

Top Papers

  1. 1
  2. 2
  3. 3

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago