Senior Motion Planning Engineer - Safety-Layer

Our teams are made up of engineers, researchers, innovators, dreamers, and doers. Together, we can change the world.

All Motional open positions and applications will be posted here on the official website.

Autonomy

Boston, Las Vegas, Pittsburgh, Remote U.S. Only

The Safety-Layer Team is responsible for ensuring the Motional Autonomous Vehicle executes appropriate dynamic trajectories in both nominal and safety-critical scenarios. This team focuses on refining upstream motion plans to guarantee compliance with safety-driven requirements, executing highly dynamic evasive maneuvers at the AV's physical limits, and implementing system-wide safety overrides. The team pursues these goals through a process of rigorous algorithmic design, implementation, validation, analysis in simulation, and comprehensive closed-course and public road evaluation.

As the team's responsibilities are growing to support more complex scenarios and the scaling of Motional’s deployment area, we are looking for a Senior Motion Planning - Safety-Layer to help build a robust motion planning safety-layer. You will work on enabling cutting-edge capabilities for Motional’s AV, including algorithm design and documentation, software development, metrics definition, simulation implementation and evaluation, and in-vehicle testing.

As part of this role, you will maintain cross-functional relationships across the organization to help understand and communicate the impacts of their work on overall AV performance.

• Design, implement, and test algorithms for synthesizing trajectories and selecting trajectories across asynchronous generation sources to ensure AV behaviors adhere to safety-critical requirement specifications.
• Develop evasive maneuver planning algorithms in conjunction with low-level control systems to safely use the full actuation capacity of Motional's AV for collision avoidance.
• Evaluate the role of multiple perception sensor modalities in refining the safety-layer's identification of safety-critical scenarios.
• Construct test/analysis tooling for isolated and system-level evaluation of safety-layer components and processes, select test parameters, and analyze test results.
• Utilize, validate, and improve dynamic simulation environments to accelerate the design and analysis of safety-critical motion planning algorithms, including validating simulation models against real-world data.
• Define and implement performance, comfort, and safety metrics for motion planning and safety systems, and perform quantitative analysis of fleet-wide vehicle data to validate design improvements and identify key areas of improvement.
• Build a system-wide understanding of Motional's autonomy stack components by supporting root-cause analysis of simulation, closed-course, and public road events from a motion planning and safety perspective.
• Travel for in-person vehicle testing at Motional's and our partner's deployment and test track locations, typically once every 4 to 6 months.

• MS, PhD preferred, in one of Control Theory / Robotics / Computer Science / Applied Math or a related field
• 3+ years of experience developing motion planning and collision avoidance related algorithms and components in an industrial setting, preferably the AV industry
• Strong background in modern C++ programming (C++17, C++20), with an emphasis on developing high-performance and reliable systems
• Experienced with concepts from motion planning including trajectory generation through numerical optimization and sampling-based approaches, cost and heuristic-based methods for trajectory selection, multi-modal prediction models, geometric and physics-based collision detection, time-series data processing, and data visualization
• Familiarity with dynamic collision modeling, convex set processing, state machine logic, and multi-rate decision-making processes
• Familiarity with agentic-based development tooling
• Familiarity with Linux-based operating systems

• General familiarity with machine learning, including supervised learning, reinforcement learning, training/test validation, etc.
• Familiarity with vehicle dynamics modeling, including standard longitudinal/lateral vehicle dynamics models
• Publications in Robotics/Controls or Autonomous Driving conferences/journals (IROS, ICRA, CoRL, RAL, TRO, CDC, ACC, TAC, CSL, Automatica, IV, ITSC, T-ITS)
• Familiarity with SQL-based queries
• Familiarity with Python
• Familiarity with Gitlab, Bazel, VSCode / CLion
• Familiarity with AV industry…