Evidential-Based Approach for Trajectory Planning With Tentacles, for Autonomous Vehicles

Mouhagir, Hafida; Talj, Reine; Cherfaoui, Véronique; Aioun, François; Guillemard, Franck

doi:10.1109/tits.2019.2930035

Cited by 18 publications

(12 citation statements)

References 42 publications

(44 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The tentacle [17] and VVF [20] algorithms were used to compare with the proposed method. These are representative and general algorithms of the candidate path selection and artificial field generation methods, respectively.…”

Section: General Model-based Motion-planning Algorithms Used For Comp...mentioning

confidence: 99%

“…Tentacle Algorithm [17] This algorithm has 16 candidate path sets depending on the velocity, and each candidate path set has 81 candidate paths. The cost for each candidate path is calculated with the objective function, and the candidate path with the smallest value is selected.…”

Section: 41mentioning

confidence: 99%

“…To select one path, the objective function is modeled to reach the global path, avoid obstacles, and keep a ride comfort. Typically, three algorithms have been used: the Dynamic window approach (DWA) [15], Curvature velocity method (CVM) [16], and tentacle [17] algorithms. The DWA algorithm designs a window according to the current state of the vehicle, and candidate paths are generated within the window.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Vision-based Autonomous Driving for Unstructured Environments Using Imitation Learning

Ahn¹,

Kim²,

Park³

2022

Preprint

View full text Add to dashboard Cite

Unstructured environments are difficult for autonomous driving. This is because various unknown obstacles are lied in drivable space without lanes, and its width and curvature change widely. In such complex environments, searching for a path in real-time is difficult. Also, inaccurate localization data reduce the path tracking accuracy, increasing the risk of collision. Instead of searching and tracking the path, an alternative approach has been proposed that reactively avoids obstacles in real-time. Some methods are available for tracking global path while avoiding obstacles using the candidate paths and the artificial potential field. However, these methods require heuristics to find specific parameters for handling various complex environments. In addition, it is difficult to track the global path accurately in practice because of inaccurate localization data. If the drivable space is not accurately recognized (i.e., noisy state), the vehicle may not smoothly drive or may collide with obstacles. In this study, a method in which the vehicle drives toward drivable space only using a vision-based occupancy grid map is proposed. The proposed method uses imitation learning, where a deep neural network is trained with expert driving data. The network can learn driving patterns suited for various complex and noisy situations because these situations are contained in the training data. Experiments with a vehicle in actual parking lots demonstrated the limitations of general model-based methods and the effectiveness of the proposed imitation learning method.

show abstract

Section: General Model-based Motion-planning Algorithms Used For Comp...mentioning

confidence: 99%

Section: 41mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Vision-based Autonomous Driving for Unstructured Environments Using Imitation Learning

Ahn¹,

Kim²,

Park³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In [9], [10] and [11], the paths are constructed as "parallel tentacles" and are defined by a lateral offset to the reference road, usually the center of the current lane. In [12], moving obstacles are considered by being extended in the underlying occupancy grid according to their heading and speed to provide some anticipation with respect to their expected motion. In these methods, the speed is not explicitly defined along the path.…”

Section: Related Workmentioning

confidence: 99%

Generic Trajectory Planning Algorithm for Urban Autonomous Driving

Duhautbout

Talj

Cherfaoui

et al. 2021

2021 20th International Conference on Advanced Robotics (ICAR)

Self Cite

View full text Add to dashboard Cite

In this paper, a new local planning algorithm for urban autonomous driving is presented. Our main contribution is to define a fully algorithmic method, based on a geometrical representation of the environment, to compute predictive speed profiles on multiple paths, to ensure safety and comfort with respect to the scene and its predicted evolution. Simulation results are provided to evaluate the behaviour of the proposed algorithm on various scenarios. Those results show a good, comfortable and safe reaction of the vehicle to its static and dynamic environment with processing times compatible with real-time control.

show abstract

“…As an important link in the construction of intelligent transportation system in the future, automatic driving is a complex of perceptual positioning, decision planning, and motion control. 1–3 SAE (Society of automotive engineers, SAE) has divided autonomous driving into six distinct levels. This definition has become a common standard adopted by many automotive industry related companies around the world.…”

Section: Introductionmentioning

confidence: 99%

Longitudinal and lateral collision avoidance control strategy for intelligent vehicles

Zhang

Wang

Zhao

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part D:

View full text Add to dashboard Cite

In order to solve the problems of longitudinal and lateral control coupling, low accuracy and poor real-time of existing control strategy in the process of active collision avoidance, a longitudinal and lateral collision avoidance control strategy of intelligent vehicle based on model predictive control is proposed in this paper. Firstly, the vehicle nonlinear coupling dynamics model is established. Secondly, considering the accuracy and real-time requirements of intelligent vehicle motion control in pedestrian crossing scene, and combining the advantages of centralized control and decentralized control, an integrated unidirectional decoupling compensation motion control strategy is proposed. The proposed strategy uses two pairs of unidirectional decoupling compensation controllers to realize the mutual integration and decoupling in both longitudinal and lateral directions. Compared with centralized control, it simplifies the design of controller, retains the advantages of centralized control, and improves the real-time performance of control. Compared with the decentralized control, it considers the influence of longitudinal and lateral control, retains the advantages of decentralized control, and improves the control accuracy. Finally, the proposed control strategy is simulated and analyzed in six working conditions, and compared with the existing control strategy. The results show that the proposed control strategy is obviously better than the existing control strategy in terms of control accuracy and real-time performance, and can effectively improve vehicle safety and stability.

show abstract

Evidential-Based Approach for Trajectory Planning With Tentacles, for Autonomous Vehicles

Cited by 18 publications

References 42 publications

Vision-based Autonomous Driving for Unstructured Environments Using Imitation Learning

Vision-based Autonomous Driving for Unstructured Environments Using Imitation Learning

Generic Trajectory Planning Algorithm for Urban Autonomous Driving

Longitudinal and lateral collision avoidance control strategy for intelligent vehicles

Contact Info

Product

Resources

About