Robust Modeling and Controls for Racing on the Edge

Spisak, Joshua; Saba, Andrew; Suvarna, Nayana; Mao, Brian; Zhang, Chuan Tian; Chang, Chiang‐Hua; Scherer, Sebastian; Ramanan, Deva

doi:10.48550/arxiv.2205.10841

Cited by 2 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where Q, R denote gain matrices for LQR. For the real-time control performance, we compute the state feedback optimal LQR gains over piecewise velocity intervals offline [21].…”

Section: Model-based Path Tracking Controlmentioning

confidence: 99%

Data-Driven Model Identification via Hyperparameter Optimization for Autonomous Racing Systems

Hyunki¹,

Chung²,

Shim³

2023

Preprint

View full text Add to dashboard Cite

In this letter, we propose a model identification method via hyperparameter optimization (MIHO). Our method adopts an efficient explore-exploit strategy to identify the parameters of dynamic models in a data-driven optimization manner. We utilize MIHO for model parameter identification of the AV-21, a full-scaled autonomous race vehicle. We then incorporate the optimized parameters for the design of model-based planning and control systems of our platform. In experiments, the learned parametric models demonstrate good fitness to given datasets and show generalization ability in unseen dynamic scenarios. We further conduct extensive field tests to validate our model-based system. The tests show that our race systems leverage the learned model dynamics and successfully perform obstacle avoidance and high-speed driving over 200km/h at the Indianapolis Motor Speedway and Las Vegas Motor Speedway. The source code for MIHO and videos of the tests are available at https://github.com/hynkis/MIHO.

show abstract

“…where Q, R denote gain matrices for LQR. For the real-time control performance, we compute the state feedback optimal LQR gains over piecewise velocity intervals offline [21].…”

Section: Model-based Path Tracking Controlmentioning

confidence: 99%

Data-Driven Model Identification via Hyperparameter Optimization for Autonomous Racing Systems

Hyunki¹,

Chung²,

Shim³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…A comprehensive overview of the current autonomous racing platforms, emphasizing the software-hardware co-evolution to the current stage, is presented in [1]. After IAC 2021 and CES 2022, each team presented its full stack of autonomous systems and approaches in [2], [3]. Moreover, the winner of the 2021 IAC, TUM Autonomous Motorsport team, is one of the pioneers of autonomous racing, sharing their state-of-the-art studies from planning [4]- [7] to optimal control systems [8], [9].…”

Section: Related Workmentioning

confidence: 99%

Resilient Navigation and Path Planning System for High-speed Autonomous Race Car

Lee¹,

Jung²,

Finazzi³

et al. 2022

Preprint

View full text Add to dashboard Cite

This paper describes a resilient navigation and planning algorithm for the high-speed Indy autonomous challenge (IAC). The IAC is a competition with full-scale autonomous race cars that drives up to 290 km/h (180 mph). However, owing to race cars' high-speed and heavy vibration, GPS/INS system is prone to degradation, causing critical localization errors and leading to serious accidents. To this end, we propose a robust navigation system to implement a multi-sensor fusion Kalman filter. We present the degradation identification based on probabilistic approaches to computing optimal measurement values for the Kalman filter correction step. Simultaneously, we present a resilient navigation system so that the race car follows the race track in the event of localization failure. In addition, an optimal path planning algorithm for obstacle avoidance is proposed. Considering the original optimal racing line, obstacles, and vehicle dynamics, we propose a roadgraph-based path planning algorithm to ensure that our race car drives in in-bounded conditions. The designed localization system was experimentally evaluated to determine its ability to handle the degraded data and prevent serious crashing accidents during high-speed driving. Finally, we describe the successful completion of the obstacle avoidance challenge.

show abstract

Robust Modeling and Controls for Racing on the Edge

Cited by 2 publications

References 11 publications

Data-Driven Model Identification via Hyperparameter Optimization for Autonomous Racing Systems

Data-Driven Model Identification via Hyperparameter Optimization for Autonomous Racing Systems

Resilient Navigation and Path Planning System for High-speed Autonomous Race Car

Contact Info

Product

Resources

About