Autonomy Engineer - Planning Under Uncertainty - Clearance Eligible - Remote
Decisive Point
Location: Remote
Time: 2 weeks ago
Responsibilities
-
- Build uncertainty-based decision-making algorithms for complex environments
- Design statistical test and evaluation pipelines for stochastic autonomy
- Develop and implement techniques to promote explainability within the decision making architecture
- Integrate with a diverse set of decision-making algorithms
- Build decision making architectures for coordination of actions across teams of homogeneous or heterogeneous autonomous agents.
- Implement, leverage and improve state-of-the-art "planning under uncertainty" algorightms for unmanned aerial systems (UAS) that to perform tactical military missions using common autonomy loops:
- Autonomy loops such as: “Sense, Make Sense, Decide, Act (SMDA)”, “Observe, Orient, Decide, Act (OODA)” loops., “Perceive, Decide, Act’ (PDA)” loops.
- Sense: Environment sensing and modeling, computer vision, sensor processing, classification, anomaly detection.
- Make Sense: Environment mapping, data interpretation, 3D voxel grids, GeoGrids, WGS84, aerospace coordinate systems and reference frames (north east down (NED), Geocentri & Geodetic latitudes, Earth-centered-inertial (ECI), earth centered earth fixed (ECEF)), no fly zones, keep-in/keep-out zones. Sensor fusion and target tracking, etc. Find, fix, track, target (F2T2).
- Decide: State machines, behavior trees, optimization algorithms, constraint solving, classic algorithms (A*, RRT*, DFS, BFS, Branch & Bound, Random Forests), heuristics, optimization, Kalman filters, particle filters, etc. Artificial intelligence techniques such as deep reinforcement learning, reinforcement learning, machine learning, neural networks, supervised learning, unsupervised learning, generic algorithms, Bayesian networks, fuzzy logic, etc.
- Act: Autonomous 2D & 3D UAS trajectory/motion planning, route planning, SLAM. Classical controls systems, optimal control systems, adaptive control systems, model predictive control systems, especially for integration of 3rd party UAS autopilots. Guidance, navigation, and controls (GNC)
- Collaborate with autonomy engineers to implemented software for autonomous UASs.
- Build, leverage, and improve robotic autonomy software architectures that can be deployed on real systems to accomplish military missions (including publish/subscribe architectures).
- Design autonomy software in collaboration with autonomy engineers that supports full integration with aircraft autopilots, datalinks, sensors, PNT/GPS/INS, ground control stations, etc.
- Support live flight test of autonomy software on F16s, group 1-5 unmanned aerial vehicles (UAVs).
- Collaborate with 3rd party UAS vehicle vendors on the integration of EpiSci autonomy software onto OEM UAS hardware.
- Collaborate with domain experts and prior DoD warfighters (ex. DoD fighter pilots) to build software autonomy solutions for military missions.
Minimum Qualifications
-
- Education Requirements:
- Bachelor’s degree in computer science/related engineering field.
- 5+ years of hands-on experience developing planning software for autonomous robotic systems that involves uncertainty.
- Technical Requirements:
- Experience developing decision making AI for autonomous unmanned systems
- Experience with software languages: Python
- Experience in one or more of the following: State Machines, Behavior Trees, Resource Allocation, Task Planning, Machine Learning, Reinforcement Learning
- Experience with probabilistic techniques: Bayesian Inference, Statistical Modeling, Probabilistic programming
- Experience developing in Docker and containerized development environments, and using Linux-based operating systems (e.g., RHEL, Ubuntu)
- Experience using git, Visual Studio Code, GitLab.
- Additional Requirements:
- Passion for solving complex problems with little supervision in a fast-moving team.
- Ability to balance multiple priorities in a fast-paced, highly collaborative, frequently changing, and sometimes ambiguous environment.
- Excellent analytical, communication, and documentation skills with demonstrated ability to collaborate across multiple teams.
- Travel Requirements
- Must be willing to travel as projects requires. Estimated average travel is once every other month for between 2 days up to 1 week. (~20%)
- Security Requirements
- Must be a U.S. Citizen.
- Must be eligible for a U.S. SECRET security clearance.
Preferred Qualifications
-
- Education Requirements:
- Master’s degree in computer science/related engineering field.
- 8+ years of experience developing multi-agent teaming software for autonomous robotic systems.
- Technical Requirements:
- Experience with modern C++.
- Experience with probabilistic programming.
- Experience working projects related to national security for one or more government agencies.
- Interdisciplinary background, with evidence of continual learning.
- Security Requirements
- An active U.S. SECRET security clearance.
EpiSci is proud to be an equal opportunity and affirmative action employer. We are committed to equal employment opportunity regardless of race, color, religion, sex, sexual orientation, age, marital status, disability, gender identity or Veteran status. We value our differences and we’re excited to learn what you can add to our team.
Don't meet every single requirement? At EpiSci, we're committed to building a diverse, inclusive, and authentic workforce, so if you're excited about this role but your previous experience doesn't align perfectly with every qualification, we encourage you to apply anyway! You may be the perfect candidate for this or another role at EpiSci.
