Staff AI/ML Vehicle Motion Control Engineer - Vehicle System Controls

Description

The Role:

The Staff AI/ML Vehicle Motion Control Engineer will be a key technical leader in GM's Vehicle System Controls organization, on a team specifically focused on AI-based control, machine learning, and advanced vehicle motion control.

You will set the technical direction for how GM combines state-of-the-art control theory with modern AI/ML methods to achieve step-change improvements in handling, comfort, safety, and efficiency. This includes classical controllers and estimators, as well as learning-based models and policies for vehicle motion control across braking, steering, chassis, and integrated dynamics.

You will collaborate closely with teams in vehicle dynamics, ADAS/AD, perception, software, and safety to architect and deliver AI-enabled motion control platforms that are robust, explainable, and production-ready.

• Technical Leadership in AI-Enabled Motion Control
  • Serve as a staff-level technical authority for AI/ML-enabled vehicle motion control within the Vehicle System Controls organization.
  • Define and own the technical roadmap for hybrid control architectures that blend model-based and data-driven methods.
  • Provide hands-on technical guidance, code and design reviews, and mentorship for engineers working in advanced control and AI/ML.
• Architecture, Algorithms, and Integration
  • Architect scalable, reusable motion control platform components and interfaces that support multiple vehicle lines and ECU/compute platforms.
  • Design and implement advanced control algorithms, including state-space control, observers/estimators, optimal/robust control, and model predictive control (MPC).
  • Integrate AI/ML components (e.g., learned models, estimators, or policies) into real-time control loops while maintaining safety, stability, and interpretability.
• AI/ML for Vehicle Motion Control and Estimation
  • Identify and lead high-value AI/ML applications in motion control, such as:
  • Develop and validate models using Python-based ML stacks (e.g., PyTorch, TensorFlow, scikit-learn, NumPy/pandas), and integrate them with embedded control software.
  • Where appropriate, apply reinforcement learning or model-based RL under safety and real-time constraints, and translate promising concepts into robust production designs
• Simulation, Data, and Tooling
  • Lead the use of MIL/SIL/HIL/DiL environments and vehicle dynamics simulation (e.g., CarSim) for development and validation of both classical and AI/ML-enabled controllers.
  • Define data workflows for collection, curation, labeling, and feature engineering from simulation, proving grounds, and fleet data to support training and validation.
  • Leverage and extend core toolchains including MATLAB/Simulink, embedded C/C++, Vehicle SPY, INCA, CANalyzer, and modern data/ML tools.
• Safety, Standards, and Productionization
  • Ensure that control and AI/ML solutions align with:
  • ISO 26262 functional safety processes and SOTIF (ISO 21448) for Safety of the Intended Functionality.
  • Emerging automotive AI safety standards and best practices, including runtime monitoring, confidence measures, and safe fallback strategies.
  • Define system-level safety concepts, monitoring logic, and fail-operational / fail-safe behaviors around AI/ML components in safety-relevant functions.
• Cross-Functional Influence and External Presence
  • Collaborate with internal stakeholders across our Milford, Michigan and Mountain View, California sites, and with external partners and academic institutions, to advance state of the art.
  • Communicate strategy, trade-offs, and technical decisions clearly to leadership, and help shape long-term investment in tools, compute, and platforms for AI in controls.

• M.Sc. or Ph.D. in Controls, Robotics, Electrical/Mechanical Engineering, Computer Engineering, Applied Mathematics, or AI/ML with focus on control, robotics, or dynamical systems.
• 8+ years of experience in control systems and embedded software development, with significant time spent on vehicle motion, chassis, or closely related dynamic systems.
• Strong foundation in control and state estimation theory and its application to real-time embedded systems, including:
• Practical experience developing and deploying embedded control software in C or C++, using MATLAB/Simulink and auto-code generation for production.
• Hands-on experience with vehicle dynamics modeling and simulation and at least one of: CarSim, similar multi-body dynamics tools, or high-fidelity in-house models.
• Proficiency with vehicle communication and measurement tools such as Vehicle SPY, INCA, and CANalyzer (or equivalent).
• Demonstrated experience using Python for data analysis and at least introductory-to-intermediate experience with machine learning or data-driven modeling applied to control, estimation, or vehicle dynamics problems.
• Proven ability to lead complex technical efforts, including roadmapping, design reviews, and mentoring of other engineers.
• Excellent communication and collaboration skills, with the ability to work effectively across disciplines and locations (Milford, Michigan and Mountain View, California).

• Deep, applied experience with AI/ML in Control, Estimation and robotics, such as:
• Data-driven dynamics modeling and system identification at scale.
  Learning-based controllers (e.g., RL, model-based RL, or approximate dynamic programming) for real systems.
  ML-based estimation and prediction for driver intent, road conditions, or environment-aware motion control.
• Applying deep learning architectures (e.g., CNNs, RNNs, and transformer-based models) to perception, estimation, or decision-making tasks that feed into vehicle motion control.
• Familiarity with large language models (LLMs) and large vision / vision-language models (e.g., LVLMs), and how their outputs can be safely incorporated into planning, diagnostics, or advanced control workflows in an automotive context.
• Experience working with modern foundation-model and multimodal AI ecosystems (e.g., tooling, prompt/response pipelines, safety filters) in conjunction with real-time or near-real-time control systems.
• Experience with modern ML engineering / MLOps practices.