plafosim.algorithms.speed_position module

class plafosim.algorithms.speed_position.FormationAlgorithm(owner: object)

Bases: ABC

Abstract base class for any type of platoon formation algorithm (i.e., assignment calculation).

Implementing sub-classes need to override the do_formation() method.

__init__(owner: object)

Initialize an instance of a formation algorithm.

Parameters:

owner (object) – The owning object that is execution this algorithm. This can be either a PlatooningVehicle or an Infrastructure.

abstract add_parser_argument_group(parser: ArgumentParser) → _ArgumentGroup

Abstract method for performing any type of platoon formation (i.e., assignment calculation).

This methods needs to be overridden in implementing sub-classes.

Returns:

The specific argument group for this algorithm.

Return type:

argparse._ArgumentGroup

abstract do_formation()

Abstract method for performing any type of platoon formation (i.e., assignment calculation).

This methods needs to be overridden in implementing sub-classes.

finish()

Reserved for future use.

_abc_impl = <_abc_data object>

property name

Print the name of the formation algorithm.

class plafosim.algorithms.speed_position.PlatoonRole(value)

Bases: Enum

A collection of available platoon roles.

FOLLOWER = 2

JOINER = 3

LEADER = 1

LEAVER = 4

NONE = 0

class plafosim.algorithms.speed_position.SpeedPosition(owner: object, alpha: float = 0.5, speed_deviation_threshold: float = 0.2, position_deviation_threshold: int = 1000, formation_centralized_kind: str = 'greedy', solver_time_limit: int = 60, record_solver_traces: bool = False, record_infrastructure_assignments: bool = False, **kw_args)

Bases: FormationAlgorithm

Platoon Formation Algorithm based on Similarity, considering Speed and Position.

See papers

Julian Heinovski and Falko Dressler, “Where to Decide? Centralized vs. Distributed Vehicle Assignment for Platoon Formation,” IEEE Transactions on Intelligent Transportation Systems, vol. 25 (11), pp. 17317–17334, November 2024. https://www.tkn.tu-berlin.de/bib/heinovski2024where/

and

Julian Heinovski and Falko Dressler, “Platoon Formation: Optimized Car to Platoon Assignment Strategies and Protocols,” Proceedings of 10th IEEE Vehicular Networking Conference (VNC 2018), Taipei, Taiwan, December 2018. https://www.tkn.tu-berlin.de/bib/heinovski2018platoon/

__init__(owner: object, alpha: float = 0.5, speed_deviation_threshold: float = 0.2, position_deviation_threshold: int = 1000, formation_centralized_kind: str = 'greedy', solver_time_limit: int = 60, record_solver_traces: bool = False, record_infrastructure_assignments: bool = False, **kw_args)

Initialize an instance of this formation algorithm to be used in a vehicle or an infrastructure.

Parameters:

• owner (object) – The owning object that is execution this algorithm. This can be either a PlatooningVehicle or an Infrastructure.

• alpha (float) – The weighting factor alpha

• speed_deviation_threshold (float) – The threshold for speed deviation

• position_deviation_threshold (int) – The threshold for position deviation

• formation_centralized_kind (str) – TODO

• solver_time_limit (int) – The time limit in s to apply to the solver

• record_solver_traces (bool) – Whether to record continuous solver traces

• record_infrastructure_assignments (bool) – Whether to record infrastructure assignments

_do_formation_centralized()

Run centralized greedy formation approach.

This selects candidates and triggers join maneuvers.

_do_formation_distributed()

Run distributed greedy formation approach.

This selects a candidate and triggers a join maneuver.

_do_formation_optimal()

Run centralized optimal formation approach.

This selects candidates and triggers join maneuvers.

_record_infrastructure_assignments(basename: str)

Record infrastructure assignments.

Parameters:

basename (str) – The basename of the result file

classmethod add_parser_argument_group(parser: ArgumentParser) → _ArgumentGroup

Create and return specific argument group for this algorithm to use in global argument parser.

Parameters:

parser (argparse.ArgumentParser) – The global argument parser

Returns:

The specific argument group for this algorithm

Return type:

argparse._ArgumentGroup

cost_speed_position(ds: float, dp: float) → float

Return the overall cost (i.e., the weighted deviation) for a candidate.

Parameters:

• ds (float) – The deviation in speed

• dp (int) – The deviation in position

Returns:

The weighted relative deviation

Return type:

float

do_formation()

Run platoon formation algorithms to search for a platooning opportunity and perform the corresponding join maneuver.

dp(vehicle: PlatooningVehicle, platoon: Platoon) → float

Return the deviation in position from a given platoon.

NOTE: In the original version of the paper, the deviation calculated here was not normalized.

Parameters:

• vehicle (PlatooningVehicle) – The vehicle for which the deviation is calculated

• platoon (Platoon) – The platoon to which the deviation is calculated

Returns:

The relative deviation in position

Return type:

float

ds(vehicle: PlatooningVehicle, platoon: Platoon) → float

Return the deviation in speed from a given platoon.

NOTE: In the original version of the paper, the deviation calculated here was not normalized.

Parameters:

• vehicle (PlatooningVehicle) – The vehicle for which the deviation is calculated

• platoon (Platoon) – The platoon to which the deviation is calculated

Returns:

The relative deviation in speed

Return type:

float

finish()

Clean up the instance of the formation algorithm.

This includes mostly statistic recording.

_abc_impl = <_abc_data object>

property name

Print the name of the formation algorithm.

plafosim.algorithms.speed_position.initialize_infrastructure_assignments(basename: str)

Initialize the infrastructure assignments result file.

Parameters:

basename (str) – The basename of the result file

plafosim.algorithms.speed_position.initialize_solver_traces(basename: str)

Initialize the solver trace file.

Parameters:

basename (str) – The basename of the trace file

plafosim.algorithms.speed_position.record_solver_trace(basename: str, step: float, iid: int, variables: int, constraints: int, run_time: float, result_status: int, solution_value: float, best_bound: float, solution_quality: float)

Record one line in the solver trace file.

Parameters:

• basename (str) – The basename of the trace file

• step (float) – The current simulation step

• iid (int) – The id of the infrastructure which executed the solver run

• variables (int) – The number of variables

• constraints (int) – The number of constraints

• run_time (float) – The run time of the solver

• result_status (int) – The result status of the solver

• solution_value (float) – The solution value of the solver

• best_bound (float) – The best bound of the problem

• solution_quality (float) – The quality of the solution

plafosim.algorithms.speed_position.strtobool(val)

Convert a string representation of truth to true (1) or false (0).

True values are ‘y’, ‘yes’, ‘t’, ‘true’, ‘on’, and ‘1’; false values are ‘n’, ‘no’, ‘f’, ‘false’, ‘off’, and ‘0’. Raises ValueError if ‘val’ is anything else.

plafosim.algorithms.speed_position.timer()

perf_counter() -> float

Performance counter for benchmarking.