REVIEW 2 major objections 5 minor 3 references
UruBots Autonomous Cars Challenge Pro Team Description Paper for FIRA 2025
T0 review · 2 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read An RC-sized autonomous car completes a 12-meter track in 28.4 seconds and avoids obstacles using a neural network trained on over 10,000 camera images.
desk verdict A competent competition team description paper whose single reported performance number is an in-sample lap time on the same track used for training. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the behavior-cloned driving policy: a neural network trained on recorded camera frames paired with human control commands, with two scalar outputs for steering and throttle. For the convolutional model, each decision comes from a single downsampled image; for the recurrent model, a few consecutive frames are fed through convolutional and LSTM layers so the decision can use motion context. At inference, the on-board computer converts the network's two outputs into pulse-width modulation signals that drive the steering servo and the motor's electronic speed controller, closing the loop at real-time rates without a separate planner.
What would settle it
Take the best trained model, place it on a fresh track of similar length but different layout, and run timed laps without any retraining: if the car leaves the course, fails to complete laps within about 30 seconds, or collides with obstacles, then the reported reliable navigation does not generalize beyond the test track. The results section gives no out-of-sample runs, so this experiment directly settles the scope of the claim.
Extended reading notes
Core claim
In the paper's own terms, the central discovery is that an RC-scale electric car can carry a complete autonomous navigation pipeline, including camera, deep network, and control signals, and drive reliably on a designed course. Two network architectures were trained and compared: a convolutional network that maps one 160x120 pixel image to steering and throttle values, and a recurrent network that maps a short temporal sequence of frames to the same controls. On a roughly 12-meter track inside a 5x5 meter area, the vehicle completed laps in an average time of 28.4 seconds at a steady speed close to 0.42 meters per second. With obstacles added, the recurrent model was more stable and reliable, which the authors attribute to its use of the temporal structure of the frames. The paper also reports precise low-speed braking and reverse testing, supporting the vehicle's readiness for the competition's second stage.
Load-bearing premise
The trained models are evaluated only on the same 12-meter track where their training data were collected, so the paper's lap-time and obstacle results rest on the assumption that the learned behavior transfers to the actual competition track without retraining.
Editorial extensions
If this is right
- A single-camera, single-board pipeline can produce closed-loop steering and throttle fast enough to complete a 12-meter lap in under 30 seconds.
- Feeding a short sequence of frames rather than one image improves obstacle handling, supporting the use of recurrent architectures for small autonomous vehicles.
- The specific drivetrain choice, a low-speed brushed motor with a servo controller, gives the fine forward and reverse control needed for the challenge's precision stage.
- A training set of over 10,000 images is sufficient for a neural policy to follow a fixed test track and respond to introduced obstacles.
Reading between the lines
- Because the test track is the same one where training data were collected, the reported lap time and obstacle avoidance are in-sample; a meaningful next step would be to run the same trained models on an unseen track with no retraining and measure lap completion and exit rate.
- The recurrent model's advantage with obstacles suggests temporal context helps near-static objects; a testable extension is whether adding simple depth cues or optical flow to the convolutional input closes most of that gap.
- The paper does not report obstacle types, speeds, or failure counts, so a concrete extension would be to publish collision rates and distance-to-obstacle statistics across repeated runs.
- If the competition track differs in lighting, wall height, or turns, the 28.4-second result is not a prediction of competition performance; sim-to-real or domain-randomized training could test whether the behavior-cloned policy transfers.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. This paper describes the hardware and software of the UruBots entry for the FIRA 2025 Autonomous Cars Challenge Pro class. The vehicle is a modified WLtoys 144001 RC chassis with a Jetson Nano, USB camera, IMU, optical encoder, and a brushed motor/ESC. The software uses the Donkey Car framework and two Keras models (a CNN and an RNN with LSTM layers) trained on over 10,000 images collected on a 12 m test track. The reported result is an average lap time of 28.4 s (about 0.42 m/s) on that track, plus a qualitative statement that the RNN avoids obstacles better than the CNN. The paper concludes that the car navigates reliably and is prepared for the competition.
Significance. The paper is a competition team description paper rather than a research article, and it should be assessed on that basis. Its strength is the concrete integration description: custom 3D-printed mounts, dual battery power, sensor fusion via an Arduino, and the use of an established open-source framework. If the performance claims were backed by out-of-sample trials and quantitative obstacle-avoidance data, the work would be a useful engineering data point for small-scale behavior cloning. However, the evaluation is entirely in-sample on the track used to collect training data, and obstacle avoidance is not quantified, so the main capability claims are not established. The paper provides a serviceable system overview but does not rigorously validate generalization.
major comments (2)
- [Sections 1.2 and 2] The training dataset of over 10,000 images was collected on the same 12 m test track on which the vehicle's average lap time of 28.4 s is measured. No train/validation split, number of runs, or evaluation on a different layout is reported. This is an in-sample evaluation: for a CNN/RNN behavior-cloning model, the reported performance could reflect memorization of the track (or, for the LSTM variant, the temporal frame order of the single loop) rather than a general path-following policy. Consequently, the Abstract's claim of navigating 'different tracks' and the Conclusion's claim of reliable navigation under FIRA-like conditions are not supported. Please add an out-of-sample evaluation on a held-out track or, at minimum, a per-run training/validation split with separate test results, and report the number of runs and the variance of the lap times.
- [Section 2] The obstacle-avoidance comparison is purely qualitative: 'the RNN model showed better performance than the CNN model' with no success rate, number of obstacle trials, or quantitative metrics such as avoidance distance, time penalty, or speed during avoidance. Since obstacle avoidance is part of the Abstract and is central to the FIRA challenge, this is a load-bearing omission. Please report the number of obstacle trials, the percentage of successful avoidances for each model, and any relevant timing or control data.
minor comments (5)
- [Section 1.2, Figure 3] The architecture description is under-specified: the number of filters, kernel sizes, and strides for the convolutional layers, the number of units in the dense layers, the sequence length for the RNN, and the training hyperparameters (optimizer, batch size, learning rate) are not given. Without these details, the CNN vs. RNN comparison is not reproducible. A table or a more detailed text description would help.
- [Section 1.2] The phrase 'the output has two thick layers' is unclear; it likely means 'two dense layers' or 'two fully connected layers'. Please correct the terminology and make the model descriptions consistent.
- [Section 2, Figure 4] The test track photo would be clearer with a scale bar and an indication of the start/finish line and the driving direction.
- [Section 2, Figure 5] The right panel is described as 'neural network activation or prediction visualization'. Please specify exactly what is plotted (e.g., a saliency map, a specific layer's activations, or the output steering command) so the reader can interpret the comparison.
- [Section 1.2 and 2] The paper does not report inference speed or frame rate on the Jetson Nano, which is important for the claim of real-time navigation. Please add these data if available.
Circularity Check
No significant circularity: the in-sample evaluation is a methodological limitation, but no result is equivalent to its inputs by construction.
full rationale
This is a team-description paper rather than a derivation chain. Its central quantitative result is the 28.4 s average lap time on a ~12 m track (Section 2), achieved by a CNN/RNN trained on 'over 10000 images and their corresponding control commands, collected during multiple runs on the test track' (Section 1.2). The evaluation track appears to be the same layout, so the reported result is in-sample and does not by itself establish generalization to 'different tracks' as claimed in the abstract. That is a real external-validity and evaluation-design limitation, but it is not circularity under the definitions used here: the lap time is not a fitted parameter, and it is not equal to the training inputs by construction. It is a measured closed-loop outcome of the trained model. The steering/throttle outputs are supervised predictions of the same kind as the training labels, which is the normal operation of behavior cloning rather than a circular derivation. The self-citations [1,2] are contextual background and are not load-bearing: no argument in the paper reduces to those references, and no uniqueness theorem or ansatz is imported from them. No equation or construction can be exhibited that forces the reported result from the input data, so no circular step is identified.
Assumptions & free parameters
free parameters (3)
- CNN/RNN model weights =
Learned from 10,000+ images
- Camera input size =
160x120 pixels
- Training epochs =
Up to 60
assumptions (3)
- domain assumption Expert demonstrations collected on the test track are sufficient to learn the driving policy.
- domain assumption The Donkey Car framework's control primitives correctly map neural network outputs to PWM signals.
- domain assumption The 12 m test track is representative of the FIRA 2025 competition tracks.
Cite this review
Pith. "Pith review of UruBots Autonomous Cars Challenge Pro Team Description Paper for FIRA 2025." pith.science (2026). https://pith.science/paper/XDHDESSG
@misc{pith2026250607348,
author = {Pith},
title = {Pith review of: UruBots Autonomous Cars Challenge Pro Team Description Paper for FIRA 2025},
year = {2026},
howpublished = {\url{https://pith.science/paper/XDHDESSG}},
note = {Machine review of arXiv:2506.07348}
}
read the original abstract
This paper describes the development of an autonomous car by the UruBots team for the 2025 FIRA Autonomous Cars Challenge (Pro). The project involves constructing a compact electric vehicle, approximately the size of an RC car, capable of autonomous navigation through different tracks. The design incorporates mechanical and electronic components and machine learning algorithms that enable the vehicle to make real-time navigation decisions based on visual input from a camera. We use deep learning models to process camera images and control vehicle movements. Using a dataset of over ten thousand images, we trained a Convolutional Neural Network (CNN) to drive the vehicle effectively, through two outputs, steering and throttle. The car completed the track in under 30 seconds, achieving a pace of approximately 0.4 meters per second while avoiding obstacles.
Figures
Figures from the paper (2 more)
Reference graph
Works this paper leans on
-
[1]
arXiv preprint arXiv:2406.08745 (2024)
Moraes, P., Peters, C., Da Rosa, A., Melgar, V., Nuñez, F., Retamar, M., Moraes, W., Saravia, V., Sodre, H., Barcelona, S., et al.: Urubots au- tonomous cars team one description paper for fira 2024. arXiv preprint arXiv:2406.08745 (2024)
arXiv 2024
-
[2]
Moraes, P., Peters, C., Sodre, H., Moraes, W., Barcelona, S., Deniz, J., Castelli, V., Guterres, B., Grando, R.: Behavior cloning for mini autonomous car path following. In: 2024 IEEE URUCON. pp. 1–5. IEEE (2024)
work page 2024
-
[3]
In: 2020 Innovations in Intelligent Systems and Applications Confer- ence (ASYU)
Pehlivan, B., Kahraman, C., Kurtel, D., Nakıp, M., Güzeliş, C.: Real-time implementation of mini autonomous car based on mobilenet-single shot de- tector. In: 2020 Innovations in Intelligent Systems and Applications Confer- ence (ASYU). pp. 1–6. IEEE (2020)
work page 2020
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Sign in to comment.