MPCC comparison, modified PPO and SAC, multi-head attention encoding and a ROS 2 training-to-control pipeline.
Official results and qualification footage. Sole builder; final time-attack placement: 14th.
Built and submitted independently.
I designed, trained, integrated, evaluated and submitted the controller. The 2025 competition stack covered the ROS 2/Gymnasium bridge, observations, steering–throttle actions, rewards and checkpoints. ONNX, TensorRT and Triton inference tooling followed in 2026.
81.32 s total · 0 collisions · 8.03 s best lap.
A separate run on a different track.
Placed ahead of entries from Autonomous Racing at Penn and the Autoware Foundation in qualification, using the official collision-adjusted standings.
Official results & full standings ↗Technical deep dive Architecture, implementation & tradeoffs
Classical optimisation × learned control
- 01
Compare
MPCC · modified PPO / SACCompare model-predictive contouring control with policy-optimisation and actor–critic variants.
- 02
Encode
multi-head attentionBuild the learned observation representation inside the private racing stack.
- 03
Integrate
ROS 2 ↔ controllerConnect observations, actions, reward signals, training and checkpoint evaluation.
- 04
Export
ONNX → TensorRT / TritonExtend policy inference with export, engine-build and serving adapters after the competition.
Controller design, representation learning and execution were part of the same independently built AutoDRIVE project.
- Controllers
- MPCC comparison · modified PPO and SAC
- Representation
- Multi-head attention encoder
- Competition
- Third in qualification · separate final time-attack result
Controller comparison & learned representation
- Compared Model Predictive Contouring Control (MPCC) with modified PPO and SAC controllers in the same autonomous-racing project.
- Built a multi-head attention encoder as part of the learned-control work.
- Developed the controller variants and encoder in a private local codebase. The retained PPO baseline uses a separate, explicit observation/action contract.
The retained PPO baseline
- Wrote the ROS 2 environment bridge around LiDAR, collision and lap feedback, with steering and throttle command publishers.
- The retained baseline includes 180 LiDAR features and 25 discrete steering–throttle combinations, with lower-throttle turning actions and higher-throttle straight actions.
- Maintained additional observation and factored-action experiments as the training interface evolved.
The training system
- Experimented with collision penalties, survival terms, throttle incentives and lap feedback. The reward defines what the agent is encouraged to do; lap outcomes establish what it actually did.
- Built checkpoint resume, evaluation and training variants around Stable-Baselines3 PPO.
- The accompanying manuscript investigates compact LiDAR policies, observation aggregation and reward shaping. I am its first author; it is not published.
The inference continuation
- In 2026, added ONNX export, TensorRT engine-build tooling and Triton serving adapters. These extensions came after the 2025 competition.
The control contract shapes the policy.
Compressing a scan changes which geometry the agent can observe. Discretising steering and throttle changes the manoeuvres it can express. Rewarding throttle commands changes the optimisation target without directly measuring speed. I treated these choices as part of the controller, alongside the network and optimiser.
Simulation competition. Third in qualification; the entry later finished 14th in the final time-attack race. The two rounds evaluate different runs and tracks.