Object Detection and Tracking for Autonomous Vehicles in Adverse Weather Conditions

Bhadoriya, Abhay Singh; Vegamoor, Vamsi; Rathinam, Sivakumar

doi:10.4271/2021-01-0079

Cited by 5 publications

(8 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Introductionmentioning

confidence: 90%

“…Portions of this journal paper have been previously published as a conference paper [11]. This work completes [11] by including validation work that was performed since the publication of the conference paper. We have also since uploaded a dataset [18] containing sensor data from lidars, five thermal cameras, a radar, IMU, and vehicle CAN bus (https://doi.org/10.15787/VTT1/B3VKEA (accessed on 14 June 2022)).…”

Section: Introductionmentioning

confidence: 90%

“…While a brief summary of our implementation is provided below, our previous work [11] contains a more detailed explanation.…”

Section: Multiple Hypothesis Trackingmentioning

confidence: 99%

“…By including this modification, the global hypothesis will always have only one track corresponding to each object tracked. Please refer to our previous conference paper [11] for more details on our MHT algorithm implementation.…”

Section: Multiple Hypothesis Trackingmentioning

confidence: 99%

“…Thermal cameras are resilient to direct sunlight. Showing the same scene as in Figure 1.Reprinted with permission from earlier conference paper Ref [11]…”

mentioning

confidence: 99%

See 4 more Smart Citations

Vehicle Detection and Tracking Using Thermal Cameras in Adverse Visibility Conditions

Bhadoriya

Vegamoor

Rathinam

2022

Sensors

Self Cite

View full text Add to dashboard Cite

Level 5 autonomy, as defined by the Society of Automotive Engineers, requires the vehicle to function under all weather and visibility conditions. This sensing problem becomes significantly challenging in weather conditions that include events such as sudden changes in lighting, smoke, fog, snow, and rain. No standalone sensor currently in the market can reliably perceive the environment in all conditions. While regular cameras, lidars, and radars will suffice for typical driving conditions, they may fail in some edge cases. The goal of this paper is to demonstrate that the addition of Long Wave Infrared (LWIR)/thermal cameras to the sensor stack on a self-driving vehicle can help fill this sensory gap during adverse visibility conditions. In this paper, we trained a machine learning-based image detector on thermal image data and used it for vehicle detection. For vehicle tracking, Joint Probabilistic Data association and Multiple Hypothesis Tracking approaches were explored where the thermal camera information was fused with a front-facing radar. The algorithms were implemented using FLIR thermal cameras on a 2017 Lincoln MKZ operating in College Station, TX, USA. The performance of the tracking algorithm has also been validated in simulations using Unreal Engine.

show abstract

Section: Introductionmentioning

confidence: 90%

Section: Introductionmentioning

confidence: 90%

“…While a brief summary of our implementation is provided below, our previous work [11] contains a more detailed explanation.…”

Section: Multiple Hypothesis Trackingmentioning

confidence: 99%

Section: Multiple Hypothesis Trackingmentioning

confidence: 99%

“…Thermal cameras are resilient to direct sunlight. Showing the same scene as in Figure 1.Reprinted with permission from earlier conference paper Ref [11]…”

mentioning

confidence: 99%

See 3 more Smart Citations

Vehicle Detection and Tracking Using Thermal Cameras in Adverse Visibility Conditions

Bhadoriya

Vegamoor

Rathinam

2022

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

RPRP-SAP: A Robust and Precise ResNet Predictor for Steering Angle Prediction of Autonomous Vehicles

Saleem,

Asim,

Van Elst

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Steering angle controller is a core module of autonomous vehicles, where a slight miscalculation can cause severe accidents. Following safety precautions, development of robust and precise steering angle predictor is an active area of research. However, existing simulator based steering angle predictors lack in predictive performance and have not been evaluated on same benchmark datasets. Furthermore, most of them are evaluated on simulated datasets and their potential on real-world data as well as in cross-domain evaluation of both types data (real-world, simulated) remain unexplored. To accelerate and expedite research related to steering angle prediction contributions of this paper are manifold: 1) It presents two benchmark datasets that are developed using Udacity and CARLA simulators. 2) Following the need for comparative study, over both simulated datasets, it benchmarks the performance of existing predictors under 2 different evaluation settings namely: same-track and cross-track evaluation. 3) In cross-domain evaluation, it explores generalization potential of predictors by training predictors on simulated data and evaluating them on 2 realworld datasets and vice versa. 4) It presents a robust and precise steering angle predictor that utilizes skip connections for proper gradient flow among different convolutional layers. In same-track evaluation where predictors are trained and evaluated on same-track data, proposed predictor outperforms existing predictors by achieving least Mean Absolute Error (MAE) of 0.13, 0.19 and 0.065 over lake track, jungle track and CARLA based datasets, respectively. Similarly, in cross-track evaluation where predictors are trained on one track and are evaluated on other track data, once again proposed predictor outperforms existing predictors by producing average least MAE errors of 0.33 and 0.06 over Udacity and CARLA datasets, respectively. Over two real-world datasets, Sully Chen and Comma.ai, the proposed predictor demonstrates superior performance compared to existing simulator-based predictors, achieving the lowest MAE of 2.41 and 0.50, respectively.

show abstract

Deep Learning-Based Object Detection in Diverse Weather Conditions

Ravinder

Jaiswal

Gulati

2022

International Journal of Intelligent Information Technologies

View full text Add to dashboard Cite

The number of different types of composite images has grown very rapidly in current years, making Object Detection, an extremely critical task that requires a deeper understanding of various deep learning strategies that help to detect objects with higher accuracy in less amount of time. A brief description of object detection strategies under various weather conditions is discussed in this paper with their advantages and disadvantages. So, to overcome this transfer learning has been used and implementation has been done with two Pretrained Models i.e., YOLO and Resnet50 with denoising which detects the object under different weather conditions like sunny, snowy, rainy, and hazy weather. And comparison has been made between the two models. The objects are detected from the images under different conditions where Resnet50 is identified to be the best Model.

show abstract

Object Detection and Tracking for Autonomous Vehicles in Adverse Weather Conditions

Cited by 5 publications

References 14 publications

Vehicle Detection and Tracking Using Thermal Cameras in Adverse Visibility Conditions

Vehicle Detection and Tracking Using Thermal Cameras in Adverse Visibility Conditions

RPRP-SAP: A Robust and Precise ResNet Predictor for Steering Angle Prediction of Autonomous Vehicles

Deep Learning-Based Object Detection in Diverse Weather Conditions

Contact Info

Product

Resources

About