PELOPS (traffic flow simulation program)

from Wikipedia, the free encyclopedia

PELOPS is a (sub) microscopic, vehicle- oriented traffic flow simulation program that is being developed by Forschungsgesellschaft Kraftfahrwesen mbh Aachen (fka) in cooperation with BMW AG . Today it is distributed and maintained by the fka.

description

The concept of PELOPS ( P rogram for E evelopment L ängsdynamischer, mikr O scopic traffic P rocesses in S ystemrelevanter environment) is the coupling in more detail submicroscopic vehicle models with microscopic traffic engineering models that allow both an investigation of the longitudinally dynamic behavior of the vehicle as well as an analysis of the traffic flow . The advantage of this method is that it can take into account all interactions between driver, vehicle and traffic. In contrast to classic simulation tools used in the automotive industry, which usually only depict a subsystem or a single, isolated overall vehicle, the approach in PELOPS therefore pursues the simulation of the three essential elements of traffic

  • Route / environment,
  • Vehicle and
  • driver

with their interactions. The elements mentioned are modeled in a modular program structure and delimited by interfaces.

Environmental model

If necessary, the environmental model allows a detailed description of the influences of a stationary traffic environment. Both the course of the road in horizontal and vertical direction over radii and transitions, as well as the number and width of the lanes are indicated. In addition to this geometric data, traffic signs and environmental conditions can be specified.

Vehicle model

In the vehicle model, the movement dynamics are calculated on the basis of the vehicle's setting parameters, such as accelerator pedal position and gear change. Since the vehicle model is represented by components and thus in great detail, parameters such as overall efficiency and consumption can also be determined with sufficient accuracy. The vehicle itself is modeled according to the cause-and-effect principle.

Driver model

The driver model represents the connection between the vehicle and traffic simulation. It is divided into a behavior and an action model. In the behavior model, the parameters of the local driving strategy are determined from the current driving state and the vehicle environment. The parameters of the local driving strategy are an acceleration desired by the driver, the lane and, if applicable, the driving gear to be engaged. Finally, in the action model, these parameters are converted into vehicle-side manipulated variables such as steering movement, pedal actuation, gear selection and setting the indicator.

The behavior model in turn consists of two parts, the following model and the lane change model. The following model describes the traffic on a one-way lane where there is no possibility to overtake or change lanes. This successor model is based on the work of Wiedemann and has been further developed in numerous points at fka over the past ten years. The lane change model covers all traffic situations that occur on multi-lane roads and in inner-city traffic. It not only includes the classic lane change situations, such as overtaking on multi-lane lanes, avoiding obstacles and changing lanes to follow a route through a road network, but also tactical considerations such as e.g. B. the flashing to provoke a letting in narrow gaps. The driver and environmental model together enable the generation of virtual traffic.

In-the-loop simulation

PELOPS can also be used as a rapid control prototyping tool in the development of driver assistance systems . Thus, when simulating new system concepts, it is no longer absolutely necessary that algorithms and functions are integrated directly into PELOPS as source code. Both hardware components ( hardware-in-the-loop , HiL) and software ( software-in-the-loop , SiL) can be operated with PELOPS in a coupled simulation. For this purpose, PELOPS provides a network connection, a serial interface that can be used, for example, to couple with dSPACE hardware, and a CAN bus . With this SiL and HiL interface it is possible to substitute each individual virtual component of the traffic with a real component. For example, a real vehicle can be driven with a real driver in virtual traffic on a test track in order to test and experience a virtual assistance system.

credentials

  1. ↑ NN: PELOPS Whitepaper (PDF; 586 kB) , Aachen 2007
  2. ↑ Benmimoun, A .; Breuer, K .; Neunzig, D .: "Analysis of the potential of assistance systems with the help of coupled CAE development tools", 19th VDI / VW joint conference - vehicle concepts for the 2nd century automotive technology, Wolfsburg 2001
  3. ↑ Benmimoun A .: “The driver as a role model for driver assistance systems? A driver model-based approach to the development of situation-adaptive ADAS “ , 13th Aachen Colloquium Vehicle and Engine Technology, Aachen 2004
  4. ^ Wiedemann, R. "Simulation of the road traffic flow", series of publications by the Institute for Transport at the University of Karlsruhe, Volume 8, Karlsruhe 1974
  5. ↑ Ehmanns, D .: "Modeling of tactical driver behavior when changing lanes", dissertation at the Institute for Automotive Engineering at RWTH Aachen University, Aachen 2002
  6. ↑ Breuer, K .: "Traffic flow simulation for the development of driver assistance systems", dissertation at the Institute for Automotive Engineering at RWTH Aachen University, Aachen 2004

Web links