[Bone AI] Autonomy and control engineer

Job Description

1) Autonomous Driving Stack Development (Perception, SLAM, Mapping, Navigation) - Design and implementation of an autonomous driving stack for aerial and ground robots: perception, SLAM/VSLAM, mapping, navigation - Development of a VSLAM/Visual Odometry pipeline and integration of the navigation stack - Development of 3D mapping/reconstruction capabilities (e.g., NeRF, Gaussian Splatting) for localization, situational awareness, and path planning - Implementation of obstacle recognition and avoidance in static and dynamic environments - Development of local path planning and obstacle avoidance behavior (e.g., pedestrians, curbs, traffic, etc.) - Implementation of path planning and scheduling for delivery scenarios (multiple waypoints, time window constraints) 2) State Estimation and Classical Control - Implementation and tuning of a state estimation and classical control system, including sensor fusion - Definition of safe control operations and interfaces that integrate with higher-level autonomous driving modules 3) Autopilot and System Integration (Firmware, Interface, Mission) API) - Integrate autonomous driving components with autopilot firmware and operating tools - Own Ardupilot/PX4 integration: custom flight mode/function development (beyond simple configuration), parameter tuning, and interface design - Design and implementation of MAVLink/ROS2 interface and mission API (control/telemetry) 4) Data Collection and Observability - Design flight data collection strategies: what to log, how to log, trigger conditions, format/storage structure - Build rapid iterative improvement loops based on flight logs/sensor data/experimental results 5) Simulation, SITL/HITL, and Regression Testing - Build and operate simulation workflows for drones and rovers (Ardupilot + Gazebo) - Build SITL/HITL pipelines, including script-based missions, adversarial scenarios, and regression testing - Support RL engineers: Improve dynamics and sensor model quality for training/evaluation 6) Cross-functional Collaboration - Collaborate with full-stack engineers to develop data links and Define video streaming requirements (latency, resolution, telemetry fields, message format) - Collaborate with RL engineers to provide a secure, low-level interface and support RL policy integration/verification.

Qualifications

- 5+ years of experience in robotics and autonomous driving (drones and/or mobile robots) - C++ and Python proficiency, understanding of real-time systems, and debugging capabilities in embedded Linux - Experience with Ardupilot, PX4, or similar flight control firmware (including custom flight mode/function development, not just simple configuration) - Understanding of MAVLink and experience with SITL workflows - Deep expertise in at least two of the following areas: - VSLAM / Visual Odometry - 3D Mapping / Reconstruction - Motion Planning and Navigation - State Estimation & Control (EKF, MPC, LQR, etc.) - Experience with ROS/ROS2 and integration of perception-planning-control modules - Ability to conduct real-world hardware experiments and rapidly improve based on flight logs - Fluent communication skills (English proficiency preferred)

Preferred Visas

Cover Letter

Optional