Global path planning for competitive robotic cars

Rizano, Tizar; Fontanelli, Daniele; Palopoli, Luigi; Pallottino, Lucia; Salaris, Paolo

doi:10.1109/cdc.2013.6760584

Cited by 11 publications

(11 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Examples include tasks such as ball-throwing [1] or trying to hit a target on a wall with a dart [2]. In other cases, individual rewards might be quantifiable for each step taken, but a global approach, considering a whole episode of execution, could yield better results, such as in autonomous racing [3].…”

Section: Introductionmentioning

confidence: 99%

Learning to Race Through Coordinate Descent Bayesian Optimisation

Oliveira

Rocha

Ott

et al. 2018

2018 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

In the automation of many kinds of processes, the observable outcome can often be described as the combined effect of an entire sequence of actions, or controls, applied throughout its execution. In these cases, strategies to optimise control policies for individual stages of the process might not be applicable, and instead the whole policy might have to be optimised at once. On the other hand, the cost to evaluate the policy's performance might also be high, being desirable that a solution can be found with as few interactions as possible with the real system. We consider the problem of optimising control policies to allow a robot to complete a given race track within a minimum amount of time. We assume that the robot has no prior information about the track or its own dynamical model, just an initial valid driving example. Localisation is only applied to monitor the robot and to provide an indication of its position along the track's centre axis. We propose a method for finding a policy that minimises the time per lap while keeping the vehicle on the track using a Bayesian optimisation (BO) approach over a reproducing kernel Hilbert space. We apply an algorithm to search more efficiently over high-dimensional policy-parameter spaces with BO, by iterating over each dimension individually, in a sequential coordinate descent-like scheme. Experiments demonstrate the performance of the algorithm against other methods in a simulated car racing environment.

show abstract

Section: Introductionmentioning

confidence: 99%

Learning to Race Through Coordinate Descent Bayesian Optimisation

Oliveira

Rocha

Ott

et al. 2018

2018 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…All these blocks are used inside a high‐level optimiser that constructs the trajectory of the carlike vehicle along a sequence of given points. ()…”

Section: Introductionmentioning

confidence: 99%

Semianalytical minimum‐time solution for the optimal control of a vehicle subject to limited acceleration

Bertolazzi

Frego

2017

Optim Control Appl Methods

View full text Add to dashboard Cite

Summary A semianalytic solution of the minimum‐time optimal control problem of a carlike vehicle is herein presented. The vehicle is subject to the effect of laminar (linear) and aerodynamic (quadratic) drag, taking into account the asymmetric bounded longitudinal accelerations. This yields a nonlinear differential equation of the Riccati kind. The associated analytic solution exists in closed form, but it is numerically ill conditioned in some situations; hence, it is reformulated using asymptotic expansions that keep the numerical computations accurate and robust in all cases. Therefore, the proposed algorithm is designed to be fast and robust in sight to be the fundamental module of a more extended optimal control problem that considers a given clothoid curve as the trajectory and the presence of a constraint on the lateral acceleration of the vehicle. The numerical stability of the computation for limit values of the parameters is essential, as shown in the numerical tests.

show abstract

“…The second category is represented by the sensor based approaches (reactive), in order to avoid the a-priori knowledge of the map and deal with unknown conditions; among them it is possible to list the Dynamic Window Approach (DWA) [10], the Velocity Obstacles approach [11], the Virtual Field Histogram (VFH) [12] and its modification VFH+ [13], the last two based on the Potential Field (PF) method [14]. It is straightforward to understand that a combination of the two categories is the best solution to the path planning problem for mobile robots and it is indeed the most adopted [15,16,17,18].…”

Section: Introductionmentioning

confidence: 99%

Avoiding local minima in the potential field method using input-to-state stability

Guerra¹,

Efimov

Zheng

et al. 2016

Control Engineering Practice

View full text Add to dashboard Cite

International audienceSupported by a novel field definition and recent control theory results, a new method to avoid local minima is proposed. It is formally shown that the system has an attracting equilibrium at the target point, repelling equilibriums in the obstacles centers and saddle points on the borders. Those unstable equilibriums are avoided capitalizing on the established Input-to-State Stability (ISS) property of this multistable system. The proposed modification of the PF method is shown to be effective by simulation for a two variables integrator and then applied to an unicycle-like wheeled mobile robots which is subject to additive input disturbances

show abstract

Global path planning for competitive robotic cars

Cited by 11 publications

References 29 publications

Learning to Race Through Coordinate Descent Bayesian Optimisation

Learning to Race Through Coordinate Descent Bayesian Optimisation

Semianalytical minimum‐time solution for the optimal control of a vehicle subject to limited acceleration

Avoiding local minima in the potential field method using input-to-state stability

Contact Info

Product

Resources

About