S-Eye

Codebase of the BSc thesis by Linfeng Guo "Integrating Explainable AI and Semantic Segmentation For Advanced Failure Prediction in Autonomous Driving"

Overview

Our tool S-Eye leverages the attention maps by the explainable AI algorithms SmoothGrad or Faster-ScoreCAM and the segmentation maps by a U-Net model. By 'Merging' both maps together road attentions and background attentions can be distinguished and turn them into a confidence score to automatically predict incoming failures of a DNN-based autonomous driving system (ADS). The figure below shows the attention maps of a confident ADS.

#incomming image and results

Dependencies

Software setup: We adopted the PyCharm Professional 2023, a Python IDE by JetBrains.

First, you need anaconda or miniconda installed on your machine. Then, you can create and install all dependencies on a dedicated virtual environment, by running one of the following commands, depending on your platform.

# macOS
conda env create -f environments.yml 

# Windows
conda env create -f windows.yml

Furthermore, for training the N-Net mode you additionally need to install Python 3.10.12 and PyTorch 2.2.1 with Torch-Vision 0.17.1.

Hardware setup: Training the DNN models (self-driving cars and U-Net) on our datasets is computationally expensive. Therefore, we recommend using a machine with a GPU. In our setting, we ran our experiments on a machine equipped with a AMD Ryzen 7 5800H CPU processor, NVIDIA GeForce RTX 3060 Laptop GPU, and 8GB DDR4 RAM.

Replicate S-Eye experiments

Datasets & Simulator

Driving datasets, self-driving car models, and simulator have a combined size of several GBs. All the Datasets can be found here: https://drive.google.com/drive/folders/18Bya6gTL_3H0UM0QG89TxCRRDfE9RiPf?usp=sharing

Mutants

We use the following mutation operators:

• udacity_add_weights_regularisation_mutated0_MP_l1_3_1
• udacity_add_weights_regularisation_mutated0_MP_l1_l2_3_2
• udacity_add_weights_regularisation_mutated0_MP_l2_3_0
• udacity_change_activation_function_mutated0_MP_exponential_4_0
• udacity_change_activation_function_mutated0_MP_hard_sigmoid_4_0
• udacity_change_activation_function_mutated0_MP_relu_4_2
• udacity_change_activation_function_mutated0_MP_selu_4_0
• udacity_change_activation_function_mutated0_MP_sigmoid_4_3
• udacity_change_dropout_rate_mutated0_MP_0.125_0.125_6_2
• udacity_change_dropout_rate_mutated0_MP_1.0_1.0_6_1
• udacity_change_activation_function_mutated0_MP_softmax_4_4
• udacity_change_activation_function_mutated0_MP_softsign_4_5
• udacity_change_activation_function_mutated0_MP_tanh_4_2
• udacity_change_dropout_rate_mutated0_MP_0.25_0.25_6_7
• udacity_change_dropout_rate_mutated0_MP_0.75_0.75_6_0
• udacity_change_label_mutated0_MP_12.5_4
• udacity_change_label_mutated0_MP_25.0_1
• udacity_change_loss_function_mutated0_MP_mean_absolute_error_2

##Instruction

• Manually changing each import Lib is needed
• For U_Net model, run 'segmentation/train'
• For XAI, run 'xai/heatmap'. Both SmoothGrad and Faster-ScoreCAM are included

Evaluation Scripts

For replicating the RQs, you can run:

• For S-Eye RAR/ARAR, run the file scripts/evaluate