Configuration

All local path planner parameters are configured in the all_settings package. This page explains available settings and tuning guidance.

Settings Structure

The local planner uses LocalOptSettings from all_settings/all_settings/all_settings.py:

class LocalOptSettings(metaclass=Settings):
    """Settings for local path planning"""
    N: int = 10
    solver: str = 'ipopt'
    solver_opts: Dict = {...}
    car_params: dict = {'l_r': 1.58/2, 'l_f': 1.58/2, 'm': 1.}
    bbox: dict = {'l': 2.7, 'w': 1.6}
    DF_MAX: float = 0.5
    ACC_MIN: float = -4.5
    ACC_MAX_FN: Union[float, Function] = 2.5
    DF_DOT_MAX: float = 1.0
    V_MIN: float = 0.0
    V_MAX: float = 10.0
    FRIC_MAX: Union[float, Function] = 7.0
    write_to_file: bool = False
    save_to_gif: bool = False
    use_history: bool = False
    filtering_method: int = 0
    distance_cone_order: bool = True

Optimization Parameters

N (Number of Waypoints)

Type: int

Default: 10

Description: Number of waypoints in the optimized trajectory.

Impact:

Larger N: Longer lookahead distance, smoother paths, but slower solve times
Smaller N: Shorter horizon, faster computation, but more myopic planning

Tuning guidance:

N=10 provides ~10-20m lookahead (depends on track spacing)
Increase if paths are too aggressive or jerky
Decrease if solver times exceed 200ms consistently
Typical range: 8-15 waypoints

solver

Type: str

Default: 'ipopt'

Options: 'ipopt', 'worhp', 'snopt'

Description: Which NLP solver to use.

Note: IPOPT is the standard choice. Other solvers require additional licenses/installations.

solver_opts

Type: Dict

Default:

{
    'ipopt.linear_solver': 'ma57',
    'expand': True,
    'ipopt.print_level': 0,
    'ipopt.sb': 'yes',
    'ipopt.ma57_automatic_scaling': 'no',
    'print_time': 0,
}

Description: Options passed to the IPOPT solver.

Key options:

ipopt.linear_solver: Which linear solver to use (ma27, ma57, ma86, ma97, mumps)
expand: Expand MX to SX expressions (typically faster)
ipopt.print_level: Verbosity (0 = silent, 5 = debug)
acceptable_tol: Convergence tolerance (default 1e-2)
acceptable_obj_change_tol: Stop when objective changes < this (default 1e-3)

Vehicle Parameters

car_params

Type: dict

Default: {'l_r': 0.79, 'l_f': 0.79, 'm': 1.0}

Description: Vehicle parameters for bicycle dynamics model.

l_r: Distance from CG to rear axle (meters)
l_f: Distance from CG to front axle (meters)
m: Vehicle mass (kg, currently unused but reserved for future dynamics)

Tuning guidance: Measure from physical vehicle. Total wheelbase = l_r + l_f = 1.58m for 2025 car.

bbox

Type: dict

Default: {'l': 2.7, 'w': 1.6}

Description: Bounding box of the vehicle (length × width in meters).

Note: Currently not actively used in collision avoidance (cone constraints are disabled), but reserved for future use.

Constraint Limits

DF_MAX

Type: float

Default: 0.5 radians (±28.6°)

Description: Maximum steering angle.

Tuning guidance: Set based on physical steering limits. Include safety margin to prevent mechanical damage.

ACC_MIN

Type: float

Default: -4.5 m/s²

Description: Maximum braking acceleration (negative value).

Tuning guidance:

Depends on tire grip and brake performance
Conservative: -3.0 m/s²
Aggressive: -5.0 to -7.0 m/s² (requires good tires and braking system)

ACC_MAX_FN

Type: float or Function

Default: 2.5 m/s²

Description: Maximum forward acceleration. Can be a constant or a function of velocity to model motor torque curves.

As constant:

ACC_MAX_FN = 2.5  # Constant acceleration limit

As function:

v = SX.sym('v')
ACC_MAX_FN = Function('acc_max', [v], [
    if_else(v < 5.0, 4.0, 2.5 - 0.1*(v-5))  # Decreases with speed
])

DF_DOT_MAX

Type: float

Default: 1.0 rad/s

Description: Maximum steering rate (how fast steering angle can change).

Tuning guidance:

Limited by steering actuator speed
Slower rates = smoother steering but slower response
Faster rates = quicker response but potential instability
Test empirically: increase until oscillations appear, then reduce 20-30%

V_MIN and V_MAX

Type: float

Default: V_MIN = 0.0, V_MAX = 10.0 m/s

Description: Velocity bounds.

Tuning guidance:

V_MIN = 0 prevents reverse driving
V_MAX sets top speed for safety
Can reduce V_MAX during testing or on tight tracks

FRIC_MAX

Type: float or Function

Default: 7.0 m/s² (~0.7g)

Description: Maximum combined acceleration (friction circle radius).

As constant:

FRIC_MAX = 7.0  # Constant friction limit

As function (velocity-dependent for aero/downforce):

v = SX.sym('v')
FRIC_MAX = Function('fric_max', [v], [
    7.0 + 0.05*v**2  # Increases with speed due to downforce
])

Tuning guidance:

Depends on tire compound, track surface, and downforce
Conservative: 6.0 m/s² (~0.6g)
Moderate: 7-8 m/s² (~0.7-0.8g)
Aggressive: 9-12 m/s² (~0.9-1.2g, requires excellent tires/aero)

Cone Ordering Configuration

distance_cone_order

Type: bool

Default: True

Description: Use distance-based greedy nearest-neighbor cone ordering algorithm.

Options:

True: Use distance-based ordering (recommended for 2025 competition)
False: Use alternative ordering algorithm (funny_cone_ordering, not fully validated)

Tuning guidance: Keep as True unless specifically testing the alternative algorithm.

Debugging and Visualization

write_to_file

Type: bool

Default: False

Description: Write path planning results to file for offline analysis.

Note: Not actively used in current implementation.

save_to_gif

Type: bool

Default: False

Description: Save cone ordering visualization frames to GIF for debugging.

Usage:

save_to_gif: bool = True

Generates local_cone_ordering.gif showing cone ordering evolution.

Warning: Significantly slows down the planner. Use only for debugging.

use_history

Type: bool

Default: False

Description: Use cone history tracking for improved ordering consistency.

Note: Experimental feature, not thoroughly tested.

filtering_method

Type: int

Default: 0

Description: Which cone filtering method to use (if any).

Note: Currently not actively used. Reserved for future filtering strategies.

ROS Topics

Subscribed Topics:

Topic	Type	Description
`/slam/map/local`	feb_msgs/Map	Local cone observations from SLAM
`/slam/map/global`	feb_msgs/Map	Global map (triggers planner shutdown)
`/slam/state`	feb_msgs/State	Current vehicle state
`/path/finished`	std_msgs/Bool	Signal from other nodes that path is complete

Published Topics:

Topic	Type	Description
`/path/local`	feb_msgs/FebPath	Optimized local trajectory
`/path/local/viz`	sensor_msgs/PointCloud	Path visualization (waypoints as points)
`/path/local/vizpath`	nav_msgs/Path	Path visualization (full poses)
`/path/local/vizconeorder`	visualization_msgs/Marker	Cone pairing visualization (red lines)

Example Configurations

Conservative (Testing)

N = 8                    # Shorter horizon
V_MAX = 5.0              # Reduced speed
ACC_MAX_FN = 1.5         # Gentle acceleration
FRIC_MAX = 5.0           # Conservative cornering
DF_DOT_MAX = 0.5         # Slow steering

Balanced (Typical Racing)

N = 10                   # Standard horizon
V_MAX = 10.0             # Moderate speed
ACC_MAX_FN = 2.5         # Balanced acceleration
FRIC_MAX = 7.0           # Good tire grip
DF_DOT_MAX = 1.0         # Quick steering response

Aggressive (Competition)

N = 12                   # Longer lookahead
V_MAX = 12.0             # High speed
ACC_MAX_FN = 3.0         # Strong acceleration
FRIC_MAX = 9.0           # High grip tires
DF_DOT_MAX = 1.5         # Fast steering

Tight/Technical Track

N = 10                   # Standard horizon
V_MAX = 8.0              # Lower speed for corners
ACC_MAX_FN = 2.0         # Moderate power
FRIC_MAX = 8.0           # Need good cornering
DF_DOT_MAX = 1.2         # Quick direction changes