Baidu Apollo EM Motion Planner

Fan, Haoyang; Zhu, Fan; Liu, Changchun; Zhang, Liangliang; Zhuang, Li; Li, Dong; Zhu, Weicheng; Hu, Jiangtao; Li, Hongye; Kong, Qi

doi:10.48550/arxiv.1807.08048

Cited by 61 publications

(87 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The fitness defines the suitability of the grouping by calculating the sum of weighted edges between different groups in Equation (6). It is designed for weighing the sum of edges in an overlapping graph, and we can therefore use it in our optimization on workload grouping.…”

Section: B Balanced Robot Data Groupingmentioning

confidence: 99%

Edge Robotics: Edge-Computing-Accelerated Multi-Robot Simultaneous Localization and Mapping

Huang¹,

Zeng²,

Xu³

et al. 2021

Preprint

View full text Add to dashboard Cite

With the wide penetration of smart robots in multifarious fields, Simultaneous Localization and Mapping (SLAM) technique in robotics has attracted growing attention in the community. Yet collaborating SLAM over multiple robots still remains challenging due to performance contradiction between the intensive graphics computation of SLAM and the limited computing capability of robots. While traditional solutions resort to the powerful cloud servers acting as an external computation provider, we show by real-world measurements that the significant communication overhead in data offloading prevents its practicability to real deployment. To tackle these challenges, this paper promotes the emerging edge computing paradigm into multi-robot SLAM and proposes RecSLAM, a multi-robot laser SLAM system that focuses on accelerating map construction process under the robot-edge-cloud architecture. In contrast to conventional multi-robot SLAM that generates graphic maps on robots and completely merges them on the cloud, RecSLAM develops a hierarchical map fusion technique that directs robots' raw data to edge servers for real-time fusion and then sends to the cloud for global merging. To optimize the overall pipeline, an efficient multi-robot SLAM collaborative processing framework is introduced to adaptively optimize robot-to-edge offloading tailored to heterogeneous edge resource conditions, meanwhile ensuring the workload balancing among the edge servers. Extensive evaluations show RecSLAM can achieve up to 39.31% processing latency reduction over the state-of-the-art. Besides, a proof-of-concept prototype is developed and deployed in real scenes to demonstrate its effectiveness.

show abstract

Section: B Balanced Robot Data Groupingmentioning

confidence: 99%

Edge Robotics: Edge-Computing-Accelerated Multi-Robot Simultaneous Localization and Mapping

Huang¹,

Zeng²,

Xu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…There is a large body of literature on the use of traffic predictions for online planning and of- fline policy learning. Analytical traffic forecasting and handcoded cost functions [17], [18] are widely used to satisfy constraints such as traffic rules and collision avoidance. On the other hand, learning-based planning methods [19]- [21] have shown progress on reactive behavior and generalization in complex driving scenarios.…”

Section: Previous Workmentioning

confidence: 99%

PredictionNet: Real-Time Joint Probabilistic Traffic Prediction for Planning, Control, and Simulation

Kamenev¹,

Bohan²,

Kulkarni³

et al. 2021

Preprint

View full text Add to dashboard Cite

Predicting the future motion of traffic agents is crucial for safe and efficient autonomous driving. To this end, we present PredictionNet, a deep neural network (DNN) that predicts the motion of all surrounding traffic agents together with the ego-vehicle's motion. All predictions are probabilistic and are represented in a simple top-down rasterization that allows an arbitrary number of agents. Conditioned on a multilayer map with lane information, the network outputs future positions, velocities, and backtrace vectors jointly for all agents including the ego-vehicle in a single pass. Trajectories are then extracted from the output. The network can be used to simulate realistic traffic, and it produces competitive results on popular benchmarks. More importantly, it has been used to successfully control a real-world vehicle for hundreds of kilometers, by combining it with a motion planning/control subsystem. The network runs faster than real-time on an embedded GPU, and the system shows good generalization (across sensory modalities and locations) due to the choice of input representation. Furthermore, we demonstrate that by extending the DNN with reinforcement learning (RL), it can better handle rare or unsafe events like aggressive maneuvers and crashes.

show abstract

“…A*, RRT, and their variations are classical approaches applied in autonomous driving [8], [9]. Considering the structured constraints of road lane, decomposing lateral and longitudinal movement planning in Frenet-frame is a promising method to simplify the optimization based planning in practice [10]. To model the inherent unpredictability of other traffic and the resulting uncertainty, a semi-reactive trajectory generation method based on polynomial approximation is proposed, which can be tightly integrated into the behavioral layer [11].…”

Section: Related Workmentioning

confidence: 99%

Reinforcement Learning based Negotiation-aware Motion Planning of Autonomous Vehicles

Wang¹,

Zhuang²,

Gu³

et al. 2021

Preprint

View full text Add to dashboard Cite

For autonomous vehicles integrating onto roadways with human traffic participants, it requires understanding and adapting to the participants' intention and driving styles by responding in predictable ways without explicit communication. This paper proposes a reinforcement learning based negotiation-aware motion planning framework, which adopts RL to adjust the driving style of the planner by dynamically modifying the prediction horizon length of the motion planner in real time adaptively w.r.t the event of a change in environment, typically triggered by traffic participants' switch of intents with different driving styles. The framework models the interaction between the autonomous vehicle and other traffic participants as a Markov Decision Process. A temporal sequence of occupancy grid maps are taken as inputs for RL module to embed an implicit intention reasoning. Curriculum learning is employed to enhance the training efficiency and the robustness of the algorithm. We applied our method to narrow lane navigation in both simulation and real world to demonstrate that the proposed method outperforms the common alternative due to its advantage in alleviating the social dilemma problem with proper negotiation skills.

show abstract

Baidu Apollo EM Motion Planner

Cited by 61 publications

References 11 publications

Edge Robotics: Edge-Computing-Accelerated Multi-Robot Simultaneous Localization and Mapping

Edge Robotics: Edge-Computing-Accelerated Multi-Robot Simultaneous Localization and Mapping

PredictionNet: Real-Time Joint Probabilistic Traffic Prediction for Planning, Control, and Simulation

Reinforcement Learning based Negotiation-aware Motion Planning of Autonomous Vehicles

Contact Info

Product

Resources

About