An innovation team for a global automotive company in San Jose, California, is looking for a postdoctoral scientist to investigate and implement algorithms that exploit vision and tactile feedback for robotic in-hand object manipulation. The focus of the research is to develop perception algorithms to estimate object states such as 6D pose and finger-object contact states such as slip that can be used by a high-level planner to perform manipulation tasks.
- Development of multi-modal perception algorithms focused on computer vision and tactile sensing for object grasping and manipulation.
- Deploy algorithms on a robotic platform and demonstrate their application on a set of tasks (e.g. in-hand repositioning of objects).
- Collaborate with local and international researchers and engineers within the company.
- File patents on developed technologies.
- Publish research results at top-tier conferences and journals in robotics as well as machine learning.
- Ph.D. in computer science, robotics, or a related field (1-3 publications)
- Experience with deep learning and deep learning platforms such as TensorFlow and PyTorch.
- Experience with temporal approaches such as LSTMs.
- Strong programming skills in C++ or Python.
- Familiarity with ROS and developing end-to-end systems.
- Experience training and testing machine learning algorithms and deploying models on a robotic platform.
- Experience with computer vision and multisensory perception is a plus.
- Experience with robotic manipulation, tactile sensing, and grasping is a plus.
- Experience with sim-to-real approaches is a plus.
Duration: 3+ years
Location: San Jose, CA (hybrid work schedule)