chemart / metabolic-robot-controller

hub:
  repo_type: generator
  builtin: metabolic-robot-controller
  tags:
  - banzhaf-yamamoto
chemistries:
- id: metabolic-robot-controller
  name: Metabolic / artificial biochemical network robot controllers
  intuition: 'A reaction network used as a brain. Sensors feed substances into the network, the network''s
    own reactions process them, and a motor consumes the concentration of a designated actuator species
    - so the robot''s behaviour *is* the network''s dynamics. Hand-designed networks do light-seeking
    and obstacle avoidance; evolved ones transfer from simulation to real hardware with little more than
    a scaling factor. The appeal is that computation and its substrate are not separable here.

    '
  origin: Ziegler & Banzhaf, 2001; Lones, Fuente, Turner et al.
  book: 16.1.3
  refs:
  - '[958]'
  - '[959]'
  - '[519]'
  - '[520]'
  - '[68]'
  - '[660]'
  family: application
  kind: generator
  constructive: false
  S:
    repr: abstract substances s1, s2, ...; sensor substances (b, c, d, e) receive sensor-driven inflow,
      the actuator substance a is read and consumed by the motor
    definition: explicit
    notes: 'The GP system that evolves these graphs is constructive (mutation and crossover add and remove
      molecules), but a single reaction graph has a fixed, enumerable species set.

      '
  R:
    definition: explicit
    arity:
    - 1
    - 2
    scheme: 'Bipartite directed weighted graph G = (V, E), V = R u S. Reaction nodes have the types s1
      -> s1'' ; s1 + s2 -> s1'' ; s1 -> s1'' + s2 ; s1 + s2 -> s1'' + s2'' (paper eqs. 4-7). Stoichiometric
      edges carry the default weight k_0; a special edge S -> R carries a catalytic or inhibitory weight
      k in [k_min, k_max]. A catalysed node becomes X -> Y [k_0] plus X + C -> Y + C [k_0 k]. The graph
      must fulfil material balance: positive species weights m with M(R) m = 0.

      '
  A:
    reactor:
    - lattice-2d
    - ode
    dilution: sensor substances flow in (eq. 30); the actuator substance is consumed at rate alpha above
      a_min (eq. 31); the 2001 lattice reactor also removes random molecules
    notes: 'Ziegler & Banzhaf (2001) simulate the graph in a 20 x 20 toroidal lattice with Moore neighbourhood;
      Ziegler, Dittrich & Banzhaf (1998) use a one-compartment ODE reactor with the same mass-action form
      for catalysis.

      '
  params:
  - name: n_substances
    type: int
    default: 8
    min: 2
    max: 500
    range: must include the input and output substances; the evolved graph of the paper's fig. 13 has
      nearly 50 nodes (molecules + reactions) and 70 edges
    role: structural
    meaning: number of molecule nodes |S|, including the sensor and actuator substances; the others are
      named s1, s2, ... (the paper numbers them; the prefix keeps '2 s2' readable)
  - name: n_reactions
    type: int
    default: 8
    min: 0
    max: 1000
    role: structural
    meaning: number of reaction nodes |R|, each kept only if the graph stays materially balanced
  - name: reaction_types
    type: list
    default:
    - 4
    - 5
    - 6
    - 7
    role: structural
    meaning: 'allowed reaction types by the paper''s equation number: 4 s1 -> s1'', 5 s1 + s2 -> s1'',
      6 s1 -> s1'' + s2, 7 s1 + s2 -> s1'' + s2'' (Table 1 uses all four)'
  - name: p_modifier
    type: float
    default: 0.5
    min: 0
    max: 1
    role: structural
    meaning: probability that a reaction node carries a catalytic or inhibitory edge (at most one per
      node, as in mutation case 1); not published
  - name: p_inhibitor
    type: float
    default: 0.5
    min: 0
    max: 1
    role: structural
    meaning: probability that a modifier edge is inhibitory rather than catalytic; not published
  - name: k_0
    type: float
    default: 1.0
    min: 0
    role: kinetic
    meaning: default weight k_default of stoichiometric edges, the rate constant of the spontaneous reaction
      (1.0 in Ziegler, Dittrich & Banzhaf 1998)
  - name: k_min
    type: float
    default: 10.0
    min: 0
    role: kinetic
    meaning: lower bound of catalytic/inhibitory weights, drawn uniformly in [k_min, k_max]
  - name: k_max
    type: float
    default: 10.0
    min: 0
    range: the 2001 paper gives no values; the default k_min = k_max = 10 is the catalytic efficiency
      kappa of the 1998 paper
    role: kinetic
    meaning: upper bound of catalytic/inhibitory weights
  - name: sensor_map
    type: dict
    default:
      b:
      - left1
      - left2
      c:
      - front1
      - front2
      d:
      - right1
      - right2
      e:
      - back1
      - back2
    role: structural
    meaning: 'input set I: each sensor substance and the Khepera proximity sensors whose maximum drives
      its inflow (eq. 29)'
  - name: sensor_readings
    type: dict
    default:
      left1: 1023
      left2: 1023
      front1: 1023
      front2: 1023
      right1: 1023
      right2: 1023
      back1: 1023
      back2: 1023
    role: population
    meaning: current sensor values in [0, 1023] by sensor name; unlisted sensors read 0. The default is
      the paper's robot surrounded by obstacles
  - name: max_inflow
    type: float
    default: 0.1
    min: 0
    max: 1
    role: population
    meaning: 'maxInflow of eq. 30: inflow of a sensor substance at maximum reading, as a fraction of the
      reactor volume per iteration (10% in the paper)'
  - name: actuator_map
    type: dict
    default:
      a: rotate
    role: structural
    meaning: 'output set O: each actuator substance and its action, one of rotate (2001: switch one motor''s
      direction above a_min), left-motor, right-motor (1998: substance drives a wheel)'
  - name: alpha
    type: float
    default: 0.9
    min: 0
    max: 1
    role: population
    meaning: 'consumption rate alpha of the actuator substance while it is above a_min (Table 2: 0.9)'
  - name: a_min
    type: float
    default: 0.1
    min: 0
    role: population
    meaning: 'actuator activation threshold a_min (Table 2: 0.1)'
  provides:
  - topology
  - stoichiometry
  - catalysts
  - rate-constants
  - mass-conservation
  - flow
  fidelity: reconstructed
  sources:
  - Ziegler, J. & Banzhaf, W. (2001). Evolving control metabolisms for a robot. Artificial Life 7(2):171-190.
    Reaction graphs sec. 3 (eqs. 4-17), material balance sec. 3.1.1, mutation sec. 3.2.1, sensors and
    actuators sec. 5.1-5.2 (eqs. 29-31), Tables 1-2. http://www.cs.mun.ca/~banzhaf/papers/metabolism.pdf
  - 'Ziegler, J., Dittrich, P. & Banzhaf, W. (1998). Towards a metabolic robot control system. In Holcombe
    & Paton (eds.), Information Processing in Cells and Tissues, pp. 305-318. Appendix, example A: mass-action
    form of catalysed reactions and k = 1.0, kappa = 10. https://users.fmi.uni-jena.de/~dittrich/p/ZDB97ipcat.ps'
  decisions:
  - 'The network is one individual of the GP system: a random reaction graph. The (mu, lambda) evolution,
    mutation, crossover, pathway extraction and the robot simulation act on many networks and are not
    generated; the v1 evolution parameter is dropped.'
  - Reaction nodes are drawn like mutation case 4 (type uniform over reaction_types, participants uniform
    over all molecules, with replacement). No-op reactions and duplicate nodes are rejected. The paper's
    'grow' initialisation is not described further.
  - Material balance (sec. 3.1.1) is tested as the existence of weights m >= 1 with M(R) m = 0, by linear
    programming. The paper's steps 4-6 set the independent weights to unity and check the rest, which
    can wrongly reject a balanced network whose positive solution needs other values. A node that would
    break the balance is rejected rather than penalised. Once the weight vector is unique, a node is kept
    iff it is orthogonal to it.
  - 'Extraction garbling: eqs. 4-7 lost their primes (printed ''s1 -> s1''); they are read as s1 -> s1'',
    etc., since the unprimed forms would be no-ops.'
  - 'Catalysis: the 2001 paper only says a catalyst multiplies the rate constant by a factor. The 1998
    appendix writes a catalysed reaction as a parallel mass-action channel w5 = k5 kappa [A][B][B] with
    k5 = k1, so a catalysed node becomes the plain reaction (k_0) plus a channel with the catalyst on
    both sides (k_0 k). The kappa glyph is lost in the extracted text and is restored from the table ''kappa
    10''.'
  - Inhibition has no published functional form. An inhibited node is one mass-action reaction with k_0
    whose rate dict records inhibitor and k_inhibition; a simulator must choose the inhibitory factor.
  - 'Modifier edges: at most one per node, as mutation case 1 implies (''adding an inhibitor/catalyst
    if none is present''). The modifier is any molecule, with k ~ U[k_min, k_max]. p_modifier and p_inhibitor
    are not in the paper.'
  - 'Rate defaults: the 2001 paper gives no k_default, k_min or k_max. The defaults k_0 = 1.0 and k_min
    = k_max = 10 are the 1998 paper''s k_i and kappa.'
  - Sensor inflow (eq. 30) is expressed per reactor iteration as a fraction of the reactor volume (cellsize
    = 1), so max_inflow = 0.1 gives 0.4 in total when all four sensor pairs read 1023, as the paper states.
    sensor_map and actuator_map replace the v1 callables.
  - 'Actuator outflow (eq. 31) is recorded as outflow alpha. The threshold gate [a] > a_min is not a rate
    law and is kept in extras.actuators. Reactor size, fill level and cycle counts (Table 2) only size
    the simulation and are dropped. No initial state is set: the paper fills 50% of the lattice without
    giving the composition.'
  - The functional set (kernel, eq. 16) and pathways P(v, w) are analyses of a graph, not part of it,
    and are not computed. extras.reaction_graph lists each reaction node with its type, modifier and the
    indices of the network reactions that implement it.
  phenomena:
  - light seeking and obstacle avoidance on a Khepera robot (1998, hand-designed enzyme-substrate network)
  - evolved networks transfer between simulation and a real Khepera, with only a scale factor on sensor
    inflow
  - the evolved kernel uses only the front and left sensor substances to raise the motor substance (fig.
    14)
  - genome length grows during evolution (introns, bloat) while the kernel stays small
  - combined genetic + metabolic + signalling networks controlling chaotic dynamical systems (Lones et
    al.)
  notes: 'The paper also defines a shortest-path ''metabolic pathway'' extraction, the subgraph P(v, w)
    with every participant of each reaction on the path, used by crossover. It would fit as a Chemart
    analysis on extras.reaction_graph.

    '