Scene Understanding for Autonomous Mobility

Honda Research Institute USA; Mountain View (CA); On-site; USD 80,000 - 120,000; Full time

Job summary

Honda Research Institute USA in Mountain View, CA is seeking a highly motivated Research Scientist to develop scene understanding technologies for autonomous mobility. The role involves researching and developing algorithms for perception while collaborating with engineers to integrate these solutions into real-world systems.

Ideal candidates should possess strong research capabilities and have experience in machine learning, particularly with PyTorch or TensorFlow. A PhD in a relevant field is required, alongside a minimum of 2 years of professional experience.

Qualifications

• PhD in computer science, robotics, or a related field.
• 2+ years of experience with machine learning and deep learning.
• Experience with frameworks such as PyTorch or TensorFlow.

Responsibilities

• Research on cutting-edge scene understanding algorithms using multimodal sensor data.
• Develop and implement machine learning and deep learning models.
• Publish scholarly articles to top-tier conferences and journals.

Skills

Machine learning

Deep learning

Social cue recognition

Semantic segmentation

Python programming

C++ programming

Education

PhD in computer science, robotics, or a related field

Tools

PyTorch

TensorFlow

Job description

Scene Understanding for Autonomous Mobility - Honda Research Institute USA

Job Number: P25F08

Honda Research Institute USA (HRI-US) is seeking a highly motivated Research Scientist to contribute scene understanding technologies for autonomous mobility. A successful candidate will focus on the research and development of scene understanding and perception algorithms and will have experience with one or more of the following: social cue recognition, semantic segmentation, action recognition, and SLAM. The ideal candidate combines strong research capabilities with practical experience deploying models in real-world systems.

Mountain View, CA

Key Responsibilities

• Research on cutting-edge scene understanding algorithms using multimodal sensor data (e.g., camera, LiDAR).
• Develop and implement machine learning and deep learning models for behavior and intent prediction, social interation understanding and important agent identification.
• Design and implement deep learning algorithms to represent semantic information as quantitative information and vice versa.
• Publish scholarly articles to top-tier conferences and journals in robotics, computer vision, and machine learning.
• Ability to analyze existing perception software stack and implement improvements on physical systems.
• Collaborate with a dynamic team of researchers and engineers to integrate the algorithms with motion planning and controls.

Minimum Qualifications

• PhD in computer science, robotics, or a related field
• 2+ years of experience with machine learning and deep learning
• Experience with frameworks such as PyTorch or TensorFlow
• Strong programming skills in Python or C++
• Experience with training, testing and fine tuning machine learning algorithms
• Experience with quantization, distillation, and other methods for improving runtime

Bonus Qualifications

• Experience with LLMs and Foundation models
• Experience in autonomous driving, robotics, or intelligent systems
• Familiarity with ROS and traffic simulators (e.g., Carla, Carmaker)
• Experience with developing end-to-end systems
• Experience with deploying models on a physical (robotic) platform
• Experience with optimization solvers