Peide Cai

Selected publications

1. 

   DQ-GAT: Towards Safe and Efficient Autonomous Driving with Deep Q-Learning and Graph Attention Networks

   Peide Cai, Hengli Wang, Yuxiang Sun, and Ming Liu

   IEEE Transactions on Intelligent Transportation Systems (T-ITS) , 2022

   Abs PDF Cite Website

   Autonomous driving in multi-agent dynamic traffic scenarios is challenging: the behaviors of road users are uncertain and are hard to model explicitly, and the ego-vehicle should apply complicated negotiation skills with them, such as yielding, merging and taking turns, to achieve both safe and efficient driving in various settings. Traditional planning methods are largely rule-based and scale poorly in these complex dynamic scenarios, often leading to reactive or even overly conservative behaviors. Therefore, they require tedious human efforts to maintain workability. Recently, deep learning-based methods have shown promising results with better generalization capability but less hand engineering efforts. However, they are either implemented with supervised imitation learning (IL), which suffers from dataset bias and distribution mismatch issues, or are trained with deep reinforcement learning (DRL) but focus on one specific traffic scenario. In this work, we propose DQ-GAT to achieve scalable and proactive autonomous driving, where graph attention-based networks are used to implicitly model interactions, and deep Q-learning is employed to train the network end-to-end in an unsupervised manner. Extensive experiments in a high-fidelity driving simulator show that our method achieves higher success rates than previous learning-based methods and a traditional rule-based method, and better trades off safety and efficiency in both seen and unseen scenarios. Moreover, qualitative results on a trajectory dataset indicate that our learned policy can be transferred to the real world for practical applications with real-time speeds.

2. 

   Vision-Based Autonomous Car Racing Using Deep Imitative Reinforcement Learning

   Peide Cai, Hengli Wang, Huaiyang Huang, Yuxuan Liu, and Ming Liu

   IEEE Robotics and Automation Letters (RA-L) &
   IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS​) , 2021

   Abs PDF Cite DOI Website

   Autonomous car racing is a challenging task in the robotic control area. Traditional modular methods require accurate mapping, localization and planning, which makes them computationally inefficient and sensitive to environmental changes. Recently, deep-learning-based end-to-end systems have shown promising results for autonomous driving/racing. However, they are commonly implemented by supervised imitation learning (IL), which suffers from the distribution mismatch problem, or by reinforcement learning (RL), which requires a huge amount of risky interaction data. In this work, we present a general deep imitative reinforcement learning approach (DIRL), which successfully achieves agile autonomous racing using visual inputs. The driving knowledge is acquired from both IL and model-based RL, where the agent can learn from human teachers as well as perform self-improvement by safely interacting with an offline world model. We validate our algorithm both in a high-fidelity driving simulation and on a real-world 1/20-scale RC-car with limited onboard computation. The evaluation results demonstrate that our method outperforms previous IL and RL methods in terms of sample efficiency and task performance.

3. 

   DiGNet: Learning Scalable Self-Driving Policies for Generic Traffic Scenarios with Graph Neural Networks

   Peide Cai, Hengli Wang, Yuxiang Sun, and Ming Liu

   IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS​) , 2021

   Abs PDF Cite Website

   Traditional modular self-driving frameworks scale poorly in new scenarios, which usually require tedious hand-tuning of rules and parameters to maintain acceptable performance in all foreseeable occasions. Therefore, robust and safe self-driving using traditional frameworks is still challenging, especially in complex and dynamic environments. Recently, deep-learning based self-driving methods have shown promising results with better generalization capability but less hand engineering effort. However, most of the previous learning-based methods are trained and evaluated in limited driving scenarios with scattered tasks, such as lane-following, autonomous braking, and conditional driving. In this paper, we propose a graph-based deep network to achieve scalable self-driving that can handle massive traffic scenarios. Specifically, more than 7,000 km of evaluation is conducted in a high-fidelity driving simulator, in which our method can obey the traffic rules and safely navigate the vehicle in a large variety of urban, rural, and highway environments, including unprotected left turns, narrow roads, roundabouts, and pedestrian-rich intersections. The results also show that our method achieves better performance over the baselines in terms of success rate.

4. 

   VTGNet: A Vision-based Trajectory Generation Network for Autonomous Vehicles in Urban Environments

   Peide Cai, Yuxiang Sun, Hengli Wang, and Ming Liu

   IEEE Transactions on Intelligent Vehicles (T-IV) , 2020

   Abs Video PDF Cite DOI Code

   Traditional methods for autonomous driving are implemented with many building blocks from perception, planning and control, making them difficult to generalize to varied scenarios due to complex assumptions and interdependencies. Recently, the end-to-end driving method has emerged, which performs well and generalizes to new environments by directly learning from export-provided data. However, many existing methods on this topic neglect to check the confidence of the driving actions and the ability to recover from driving mistakes. In this paper, we develop an uncertainty-aware end-to-end trajectory generation method based on imitation learning. It can extract spatiotemporal features from the front-view camera images for scene understanding, and then generate collision-free trajectories several seconds into the future. The experimental results suggest that under various weather and lighting conditions, our network can reliably generate trajectories in different urban environments, such as turning at intersections and slowing down for collision avoidance. Furthermore, closed-loop driving tests suggest that the proposed method achieves better cross-scene/platform driving results than the state-of-the-art (SOTA) end-to-end control method, where our model can recover from off-center and off-orientation errors and capture 80% of dangerous cases with high uncertainty estimations.

5. 

   Probabilistic End-to-End Vehicle Navigation in Complex Dynamic Environments With Multimodal Sensor Fusion

   Peide Cai, Sukai Wang, Yuxiang Sun, and Ming Liu

   IEEE Robotics and Automation Letters (RA-L) &
   IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS​) , 2020

   Abs Video PDF Cite DOI Website

   All-day and all-weather navigation is a critical capability for autonomous driving, which requires proper reaction to varied environmental conditions and complex agent behaviors. Recently, with the rise of deep learning, end-to-end control for autonomous vehicles has been studied a lot. However, most works are solely based on visual information, which could be degraded by challenging illumination conditions, such as dim lights or total darkness. In addition, they usually generate and apply deterministic control commands without considering the uncertainties in the future. In this paper, based on imitation learning, we propose a probabilistic driving model with multi-perception capability utilizing the information from camera, lidar and radar. We further evaluate its driving performance on-line on our new driving benchmark, which includes various environments (urban and rural areas, weathers, time of day, etc) and dynamic obstacles (vehicles, pedestrians, motorcyclists, bicyclists, etc). The results suggest that our proposed model outperforms baselines and achieves excellent generalization performance in unseen environments with heavy traffic and extreme weather conditions.

6. 

   High-Speed Autonomous Drifting With Deep Reinforcement Learning

   Peide Cai*, Xiaodong Mei*, Lei Tai, Yuxiang Sun, and Ming Liu

   IEEE Robotics and Automation Letters (RA-L) &
   IEEE International Conference on Robotics and Automation (ICRA) , 2020

   Abs Video PDF Cite DOI Code

   Drifting is a complicated task for autonomous vehicle control. Most traditional methods in this area are based on motion equations derived by the understanding of vehicle dynamics, which is difficult to be modeled precisely. We propose a robust drift controller without explicit motion equations, which is based on the latest model-free deep reinforcement learning algorithm soft actor-critic. The drift control problem is formulated as a trajectory following task, where the error-based state and reward are designed. After being trained on tracks with different levels of difficulty, our controller is capable of making the vehicle drift through various sharp corners quickly and stably in the unseen map. The proposed controller is further shown to have excellent generalization ability, which can directly handle unseen vehicle types with different physical properties, such as mass, tire friction, etc.

7. 

   Vision-Based Trajectory Planning via Imitation Learning for Autonomous Vehicles

   Peide Cai, Yuxiang Sun, Yuying Chen, and Ming Liu

   In IEEE Intelligent Transportation Systems Conference (ITSC) , 2019

   Abs Video PDF Cite DOI

   Reliable trajectory planning like human drivers in real-world dynamic urban environments is a critical capability for autonomous driving. To this end, we develop a vision and imitation learning-based planner to generate collision-free trajectories several seconds into the future. Our network consists of three sub-networks to conduct three basic driving tasks: keep straight, turn left and turn right. During the planning process, high-level commands are received as prior information to select a specific sub-network. We create our dataset from the Robotcar dataset, and the experimental results suggest that our planner is able to reliably generate trajectories in various driving tasks, such as turning at different intersections, lane-keeping on curved roads and changing lanes for collision avoidance.

Service