Robotics Researcher – Navigation

We're building the systems that let an advanced humanoid robotics platform understand where it is and decide where to go. As a Robotics Researcher in Navigation, you'll own the full stack from state estimation and mapping through global and local planning, closing the loop between perception outputs and motion execution in dynamic, unstructured real-world environments. This is an applied research role. You'll train and deploy in simulation, validate on physical hardware, and iterate until it works in the real world.

What you'll do

• Research, develop and deploy navigation algorithms for bipedal humanoid robots in complex indoor environments
• Own localisation, mapping and SLAM pipelines running in real time on embedded compute
• Build planning frameworks accounting for legged-platform constraints: footstep planning, terrain traversal, dynamic obstacle avoidance
• Integrate navigation with the broader autonomy stack including perception and locomotion
• Develop data collection and evaluation infrastructure to benchmark performance across environments
• Systematically close the sim-to-real gap through simulation and hardware iteration
• Contribute to open-source navigation research and tooling

What you'll bring

• Strong foundations in estimation theory, probabilistic robotics and motion planning
• Proven track record building and deploying SLAM, planning or autonomous navigation systems on real robots
• Strong Python and C\+\+; familiarity with ROS or equivalent middleware
• Experience getting systems to work end-to-end, not just in simulation
• Ability to debug across the hardware-software boundary and move fast on ambiguous problems

Nice to have

• Prior work on legged or humanoid navigation specifically
• Learning-based planning or visuomotor navigation policies
• Neural map representations or semantic scene understanding
• Publications at ICRA, IROS, CoRL, RSS or equivalent