Nicolas Gorlo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Nicolas Gorlo has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Luca Carlone
Massachusetts Institute of Technology
2 shared publications - Dominic Maggio
Massachusetts Institute of Technology Cambridge
1 shared publication - Eugenio Cuniato
ETH Zurich
1 shared publication - Gabriel Käppeli
Autonomous Systems Lab, ETH Zürich, Zürich, Switzerland
1 shared publication - Henriette Stadler
Autonomous Systems Lab, ETH Zürich, Zürich, Switzerland
1 shared publication - Hua Shen
Autonomous Systems Lab, ETH Zürich, Zürich, Switzerland
1 shared publication
Latest Publications
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (12) , 14140-14147
FOUND-IT: Foundation-Model-First Task-Driven 3D Scene Graphs With Granularity on Demand
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (12) , 14140-14147
FOUND-IT: Foundation-Model-First Task-Driven 3D Scene Graphs With Granularity on Demand
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (12) , 10978-10985
Long-Term Human Trajectory Prediction Using 3D Dynamic Scene Graphs
We present a novel approach for long-term human trajectory prediction in indoor human-centric environments, which is essential for long-horizon robot planning in these environments. State-of-the-art human trajectory prediction methods are limited by their focus on collision avoidance and shortterm planning, and their inability to model complex interactions of humans with the environment. In contrast, our approach overcomes these limitations by predicting sequences of human interactions with the environment and using this information to guide trajectory predictions over a horizon of up to 60 s. We leverage Large Language Models (LLMs) to predict interactions with the environment by conditioning the LLM prediction on rich contextual information about the scene. This information is given as a 3D Dynamic Scene Graph that encodes the geometry, semantics, and traversability of the environment into a hierarchical representation. We then ground these interaction sequences into multi-modal spatio-temporal distributions over human positions using a probabilistic approach based on continuous-time Markov Chains. To evaluate our approach, we introduce a new semisynthetic dataset of long-term human trajectories in complex indoor environments, which also includes annotations of humanobject interactions. We show in thorough experimental evaluations that our approach achieves a 54% lower average negative log-likelihood and a 26.5% lower Best-of-20 displacement error compared to the best non-privileged (i.e., evaluated in a zero-shot fashion on the dataset) baselines for a time horizon of 60 s
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (12) , 10978-10985
Long-Term Human Trajectory Prediction Using 3D Dynamic Scene Graphs
We present a novel approach for long-term human trajectory prediction in indoor human-centric environments, which is essential for long-horizon robot planning in these environments. State-of-the-art human trajectory prediction methods are limited by their focus on collision avoidance and shortterm planning, and their inability to model complex interactions of humans with the environment. In contrast, our approach overcomes these limitations by predicting sequences of human interactions with the environment and using this information to guide trajectory predictions over a horizon of up to 60 s. We leverage Large Language Models (LLMs) to predict interactions with the environment by conditioning the LLM prediction on rich contextual information about the scene. This information is given as a 3D Dynamic Scene Graph that encodes the geometry, semantics, and traversability of the environment into a hierarchical representation. We then ground these interaction sequences into multi-modal spatio-temporal distributions over human positions using a probabilistic approach based on continuous-time Markov Chains. To evaluate our approach, we introduce a new semisynthetic dataset of long-term human trajectories in complex indoor environments, which also includes annotations of humanobject interactions. We show in thorough experimental evaluations that our approach achieves a 54% lower average negative log-likelihood and a 26.5% lower Best-of-20 displacement error compared to the best non-privileged (i.e., evaluated in a zero-shot fashion on the dataset) baselines for a time horizon of 60 s
- 8 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics & Automation Magazine
- Vol. 31 (2) , 66-77
Geranos: A Novel Tilted-Rotors Aerial Robot for the Transportation of Poles
In challenging terrains, constructing structures, such as antennas and cable car masts, often requires the use of helicopters to transport loads via ropes. The swinging of the loads, exacerbated by wind, impairs positioning accuracy, therefore necessitating precise manual placement by ground crews. This increases costs and the risk of injuries. Challenging this paradigm, we present Geranos, a specialized multirotor unmanned aerial vehicle (UAV) designed to enhance aerial transportation and assembly. Geranos demonstrates exceptional prowess in accurately positioning vertical poles, achieving this through an innovative integration of load transport and precision. Its unique ring design mitigates the impact of high pole inertia, while a lightweight two-part grasping mechanism ensures secure load attachment without active force. With four primary propellers countering gravity and four auxiliary ones enhancing lateral precision, Geranos achieves comprehensive position and attitude control around hovering. Our experimental demonstration mimicking antenna/cable car mast installations showcases Geranos’ ability in stacking poles (3 kg, 2 m long) with remarkable sub-5-cm placement accuracy without the need of human manual intervention.
- 8 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics & Automation Magazine
- Vol. 31 (2) , 66-77
Geranos: A Novel Tilted-Rotors Aerial Robot for the Transportation of Poles
In challenging terrains, constructing structures, such as antennas and cable car masts, often requires the use of helicopters to transport loads via ropes. The swinging of the loads, exacerbated by wind, impairs positioning accuracy, therefore necessitating precise manual placement by ground crews. This increases costs and the risk of injuries. Challenging this paradigm, we present Geranos, a specialized multirotor unmanned aerial vehicle (UAV) designed to enhance aerial transportation and assembly. Geranos demonstrates exceptional prowess in accurately positioning vertical poles, achieving this through an innovative integration of load transport and precision. Its unique ring design mitigates the impact of high pole inertia, while a lightweight two-part grasping mechanism ensures secure load attachment without active force. With four primary propellers countering gravity and four auxiliary ones enhancing lateral precision, Geranos achieves comprehensive position and attitude control around hovering. Our experimental demonstration mimicking antenna/cable car mast installations showcases Geranos’ ability in stacking poles (3 kg, 2 m long) with remarkable sub-5-cm placement accuracy without the need of human manual intervention.