Autonomy Robotics Intern: Real-World SLAM & Mapping

Bedrock Robotics Inc; San Francisco (CA); On-site; USD 42,000 - 66,000; Full time

Job summary

Bedrock Robotics Inc. is seeking an internship on its State Estimation team to advance maps and localization for autonomous excavators. You’ll explore learning-based methods to improve a geometry-first mapping stack and test ideas on real fleet data.

The role involves prototyping SLAM, fusing lidar/camera data, and delivering a practical prototype with experimental results for team review. Collaboration with perception and planning is essential.

Qualifications

• Pursuing a BS, MS, or PhD in CS, robotics, electrical engineering, or related field.
• Strong Python skills and hands-on training with PyTorch.
• Solid understanding of 3D geometry and transforms.
• Familiarity with SLAM, mapping fundamentals, point clouds, or depth data.
• Comfort with messy sensor data and designing experiments.

Responsibilities

• Prototype learned SLAM and mapping methods, such as place recognition, odometry, depth completion, and neural occupancy or surface representations.
• Fuse lidar and camera segmentation into consistent 3D semantic maps, potentially using vision foundation models or open-vocabulary segmentation.
• Develop methods that handle changing terrain, moving material, sparse returns, dust, occlusion, and perceptual aliasing.
• Train models on fleet lidar and camera data, and build evaluation pipelines to compare mapping and localization performance against existing methods and ground truth.
• Work with perception and planning teams to identify the map properties that matter most for downstream decisions.
• Deliver a documented prototype, experimental results, and recommendations for future work

Skills

Python

PyTorch

3D geometry

SLAM

Experiment design

Education

BS/MS/PhD candidate

Tools

ROS

C++

Job description

Bedrock Robotics Inc. is seeking an internship on its State Estimation team to advance maps and localization for autonomous excavators. You’ll explore learning-based methods to improve a geometry-first mapping stack and test ideas on real fleet data.

The role involves prototyping SLAM, fusing lidar/camera data, and delivering a practical prototype with experimental results for team review. Collaboration with perception and planning is essential.