An integrated algorithm for ego-vehicle and obstacles state estimation for autonomous driving

Bersani, Mattia; Mentasti, Simone; Dahal, Pragyan; Arrigoni, Stefano; Vignati, Michele; Cheli, Federico; Matteucci, Matteo

doi:10.1016/j.robot.2020.103662

Cited by 33 publications

(10 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ego-vehicle state and obstacles positions and velocities are directly computed by the simulator and assumed as known. In [53] a real implementation of and estimator for these quantities based on a kalman filter algorithm that combines lidars and radars data is presented. Simulations are performed on a laptop featuring an Intel i7-3610QM CPU and 8 GB of RAM.…”

Section: Simulation Resultsmentioning

confidence: 99%

NMPC trajectory planner for urban autonomous driving

Micheli,

Bersani,

Arrigoni

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents a trajectory planner for autonomous driving based on a Nonlinear Model Predictive Control (NMPC) algorithm that accounts for Pacejka's nonlinear lateral tyre dynamics as well as for zero speed conditions through a novel slip angles calculation. In the NMPC framework, road boundaries and obstacles (both static and moving) are taken into account thanks to soft and hard constraints implementation. The numerical solution of the NMPC problem is carried out using ACADO toolkit coupled with the quadratic programming solver qpOASES. The effectiveness of the proposed NMPC trajectory planner has been tested using CarMaker multibody models. Time analysis results provided by the simulations shown, state that the proposed algorithm can be implemented on the real-time control framework of an autonomous vehicle under the assumption of data coming from an upstream estimation block.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

NMPC trajectory planner for urban autonomous driving

Micheli,

Bersani,

Arrigoni

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The first two are natural evolutions of the robotics approach; projection methods perform a projection of the 3D pointcloud on a 2D plane and then proceed to identify obstacles on the 2D grid. Two class of solution has been identified for this processing phase; the first one is based on geometry and computer vision [13], [8]. While the second one leverages on the increased available computational power, employing deep learning techniques to process the 2D grid with convolutional neural network [14], [15].…”

Section: Related Workmentioning

confidence: 99%

“…Despite using similar sensor suits, autonomous vehicles and logistics robots have some significant differences. Indoor robots can navigate the environment with low-level representations such as grid-map [7]; contrarily, planning algorithms for autonomous vehicles, which move at high speed in dynamic environments, require a more highlevel representation, like a list of 3D bounding boxes [8]. Because of these differences, it is impossible to directly employ classical robotics solutions, which work well with limited plane lidars, on autonomous vehicles.…”

Section: Introductionmentioning

confidence: 99%

Two algorithms for vehicular obstacle detection in sparse pointcloud

Mentasti¹,

Matteucci²,

Arrigoni³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

One of the main components of an autonomous vehicle is the obstacle detection pipeline. Most prototypes, both from research and industry, rely on lidars for this task. Pointcloud information from lidar is usually combined with data from cameras and radars, but the backbone of the architecture is mainly based on 3D bounding boxes computed from lidar data. To retrieve an accurate representation, sensors with many planes, e.g., greater than 32 planes, are usually employed. The returned pointcloud is indeed dense and well defined, but high-resolution sensors are still expensive and often require powerful GPUs to be processed. Lidars with fewer planes are cheaper, but the returned data are not dense enough to be processed with state of the art deep learning approaches to retrieve 3D bounding boxes. In this paper, we propose two solutions based on occupancy grid and geometric refinement to retrieve a list of 3D bounding boxes employing lidar with a low number of planes (i.e., 16 and 8 planes). Our solutions have been validated on a custom acquired dataset with accurate ground truth to prove its feasibility and accuracy.

show abstract

“…In particular, the solution adopted makes use of the road reference frame both for obstacle tracking and ego vehicle state estimation instead of a more traditional Cartesian reference frame on which sensors provide data. This allows an accurate estimate of the position of each obstacle and ego vehicle in the same reference frame used by the planning algorithm, removing the need for an intermediate conversion which would introduce approximation errors and delays [15]. A common approach in literature to deal with the overall driving problem is to decouple it into a sequence of different tasks by means of a hierarchical structure composed of different layers [16], [17] as reported in fig.…”

Section: Software Architecturementioning

confidence: 99%

Design of a prototypical platform for autonomous and connected vehicles

Arrigoni,

Mentasti,

Cheli

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Self-driving technology is expected to revolutionize different sectors and is seen as the natural evolution of road vehicles. In the last years, real-world validation of designed and virtually tested solutions is growing in importance since simulated environments will never fully replicate all the aspects that can affect results in the real world. To this end, this paper presents our prototype platform for experimental research on connected and autonomous driving projects. In detail, the paper presents the overall architecture of the vehicle focusing both on mechanical aspects related to remote actuation and sensors set-up and software aspects by means of a comprehensive description of the main algorithms required for autonomous driving as ego-localization, environment perception, motion planning, and actuation. Finally, experimental tests conducted in an urban-like environment are reported to validate and assess the performances of the overall system.

show abstract

An integrated algorithm for ego-vehicle and obstacles state estimation for autonomous driving

Cited by 33 publications

References 49 publications

NMPC trajectory planner for urban autonomous driving

NMPC trajectory planner for urban autonomous driving

Two algorithms for vehicular obstacle detection in sparse pointcloud

Design of a prototypical platform for autonomous and connected vehicles

Contact Info

Product

Resources

About