tracktor 0.4.1

Multi-target tracking with random finite sets
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202

//! Multi-Sensor LMB Tracking Example
//!
//! Demonstrates multi-sensor fusion using different strategies:
//! - AA-LMB: Arithmetic Average
//! - GA-LMB: Geometric Average (Covariance Intersection)
//! - PU-LMB: Parallel Update (Information Filter)
//! - IC-LMB: Iterated Corrector

use tracktor::filters::lmb::{
    LabeledBirthModel, LmbTrack, MultisensorLmbFilterBuilder, SensorConfig, extract_lmb_estimates,
};
use tracktor::prelude::*;

// Birth model for multi-sensor scenario
#[derive(Clone)]
struct MultisensorBirthModel;

impl LabeledBirthModel<f64, 4> for MultisensorBirthModel {
    fn birth_tracks(&self, label_gen: &mut LabelGenerator) -> Vec<BernoulliTrack<f64, 4>> {
        let cov = StateCovariance::from_matrix(nalgebra::matrix![
            200.0, 0.0, 0.0, 0.0;
            0.0, 200.0, 0.0, 0.0;
            0.0, 0.0, 50.0, 0.0;
            0.0, 0.0, 0.0, 50.0
        ]);

        vec![
            BernoulliTrack::new(
                label_gen.next_label(),
                0.02,
                GaussianState::new(1.0, StateVector::from_array([0.0, 0.0, 0.0, 0.0]), cov),
            ),
            BernoulliTrack::new(
                label_gen.next_label(),
                0.02,
                GaussianState::new(1.0, StateVector::from_array([100.0, 100.0, 0.0, 0.0]), cov),
            ),
        ]
    }

    fn expected_birth_count(&self) -> f64 {
        0.04
    }
}

fn main() {
    println!("Multi-Sensor LMB Filter Comparison");
    println!("===================================\n");

    // Shared models
    let transition = ConstantVelocity2D::new(2.0, 0.98);
    // Sensor configurations (different noise levels)
    let sensor1_obs = PositionSensor2D::new(8.0, 0.85); // Noisier sensor
    let sensor1_clutter = UniformClutter2D::new(3.0, (-50.0, 150.0), (-50.0, 150.0));

    let sensor2_obs = PositionSensor2D::new(5.0, 0.92); // Better sensor
    let sensor2_clutter = UniformClutter2D::new(2.0, (-50.0, 150.0), (-50.0, 150.0));

    let sensor_configs = vec![
        SensorConfig::new(sensor1_obs.clone(), sensor1_clutter.clone(), 0.4),
        SensorConfig::new(sensor2_obs.clone(), sensor2_clutter.clone(), 0.6),
    ];

    // Build AA-LMB filter
    let aa_filter = MultisensorLmbFilterBuilder::new(transition.clone(), MultisensorBirthModel)
        .with_sensor_weights(vec![0.4, 0.6])
        .build_aa(10);

    // Build GA-LMB filter
    let ga_filter = MultisensorLmbFilterBuilder::new(transition.clone(), MultisensorBirthModel)
        .with_sensor_weights(vec![0.4, 0.6])
        .build_ga();

    // Build IC-LMB filter
    let ic_filter = MultisensorLmbFilterBuilder::new(transition.clone(), MultisensorBirthModel)
        .with_sensor_weights(vec![0.4, 0.6])
        .build_ic();

    // Initial track
    let make_initial_tracks = || {
        vec![LmbTrack::new(
            Label::new(0, 0),
            0.9,
            GaussianState::new(
                1.0,
                StateVector::from_array([50.0, 50.0, 3.0, 2.0]),
                StateCovariance::from_matrix(nalgebra::matrix![
                    20.0, 0.0, 0.0, 0.0;
                    0.0, 20.0, 0.0, 0.0;
                    0.0, 0.0, 10.0, 0.0;
                    0.0, 0.0, 0.0, 10.0
                ]),
            ),
        )]
    };

    let mut aa_state = aa_filter.initial_state();
    aa_state.tracks.tracks = make_initial_tracks();
    let mut ga_state = ga_filter.initial_state();
    ga_state.tracks.tracks = make_initial_tracks();
    let mut ic_state = ic_filter.initial_state();
    ic_state.tracks.tracks = make_initial_tracks();

    // Simulated measurements from both sensors
    let sensor1_measurements = [
        vec![[53.0, 52.0], [120.0, 80.0]], // Detection + clutter
        vec![[57.0, 55.0]],                // Detection only
        vec![[60.0, 57.0], [10.0, 90.0]],  // Detection + clutter
    ];

    let sensor2_measurements = [
        vec![[52.0, 51.0]],               // Detection only
        vec![[55.0, 53.0], [80.0, 30.0]], // Detection + clutter
        vec![[59.0, 56.0]],               // Detection only
    ];

    let dt = 1.0;

    println!("Running 3 time steps with 2 sensors...\n");

    for t in 0..3 {
        let s1_meas: Vec<Measurement<f64, 2>> = sensor1_measurements[t]
            .iter()
            .map(|m| Measurement::from_array(*m))
            .collect();
        let s2_meas: Vec<Measurement<f64, 2>> = sensor2_measurements[t]
            .iter()
            .map(|m| Measurement::from_array(*m))
            .collect();

        let all_measurements = vec![s1_meas, s2_meas];

        println!("Time {}", t);
        println!(
            "  Sensor 1: {} meas, Sensor 2: {} meas",
            all_measurements[0].len(),
            all_measurements[1].len()
        );

        // Run all filters
        let (aa_updated, _) = aa_filter.step(aa_state, &all_measurements, &sensor_configs, dt);
        let (ga_updated, _) = ga_filter.step(ga_state, &all_measurements, &sensor_configs, dt);
        let (ic_updated, _) = ic_filter.step(ic_state, &all_measurements, &sensor_configs, dt);

        aa_state = aa_updated;
        ga_state = ga_updated;
        ic_state = ic_updated;

        // Prune
        aa_state.tracks.prune_by_existence(0.01);
        ga_state.tracks.prune_by_existence(0.01);
        ic_state.tracks.prune_by_existence(0.01);

        // Extract estimates
        let aa_est = extract_lmb_estimates(&aa_state);
        let ga_est = extract_lmb_estimates(&ga_state);
        let ic_est = extract_lmb_estimates(&ic_state);

        println!(
            "  AA-LMB: {} tracks, exp={:.2}",
            aa_state.num_tracks(),
            aa_state.expected_target_count()
        );
        if let Some(e) = aa_est.first() {
            println!(
                "    Best: ({:.1}, {:.1}), r={:.3}",
                e.state.index(0),
                e.state.index(1),
                e.existence
            );
        }

        println!(
            "  GA-LMB: {} tracks, exp={:.2}",
            ga_state.num_tracks(),
            ga_state.expected_target_count()
        );
        if let Some(e) = ga_est.first() {
            println!(
                "    Best: ({:.1}, {:.1}), r={:.3}",
                e.state.index(0),
                e.state.index(1),
                e.existence
            );
        }

        println!(
            "  IC-LMB: {} tracks, exp={:.2}",
            ic_state.num_tracks(),
            ic_state.expected_target_count()
        );
        if let Some(e) = ic_est.first() {
            println!(
                "    Best: ({:.1}, {:.1}), r={:.3}",
                e.state.index(0),
                e.state.index(1),
                e.existence
            );
        }
        println!();
    }
}