0

I have a continuous int array of polar points and I want to plot it in 3D. I heard about mplot3d and point cloud localization but I don't know how to use/implement it in my code. 

"[3085, 3084, 3075, 3068, 3069, 3051, 3049, 3042, 3037, 3039, 3028, 3024, 3016, 3014, 3004, 3002, 2991, 2989, 2983, 2976, 2974, 2972, 2967, 2956, 2956, 2955, 2945, 2942, 2943, 2932, 2934, 2925, 2924, 2916, 2911, 2909, 2905, 2895, 2894, 2895, 2888, 2888, 2881, 2878, 2877, 2870, 2871, 2866, 2870, 2862, 2854, 2853, 2854, 2851, 2844, 2840, 2839, 2843, 2837, 2831, 2831, 2832, 2829, 2827, 2821, 2819, 2824, 2822, 2816, 2819, 2815, 2812, 2807, 2812, 2812, 2811, 2800, 2804, 2803, 2801, 2802, 2798, 2796, 2793, 2799, 2793, 2795, 2796, 2792, 2792, 2786, 2787, 2790, 2787, 2783, 2788, 2790, 2789, 2781, 2788, 2782, 2784, 2785, 2787, 2780, 2790, 2789, 2784, 2793, 2789, 2787, 2789, 2787, 2791, 2790, 2791, 2794, 2793, 2790, 2799, 2799, 2795, 2802, 2797, 2797, 2801, 2806, 2801, 2804, 2808, 2807, 2806, 2809, 2815, 2813, 2817, 2815, 2819, 2821, 2820, 2826, 2826, 2829, 2833, 2835, 2839, 2836, 2839, 2842, 2844, 2844, 2852, 2852, 2850, 2860, 2861, 2867, 2871, 2871, 2877, 2875, 2877, 2883, 2890, 2893, 2889, 2899, 2901, 2907, 2909, 2912, 2918, 2926, 2925, 2935, 2931, 2940, 2936, 2939, 2954, 2954, 2965, 2963, 2971, 2979, 2982, 2979, 2944, 2907, 2882, 2844, 2815, 2783, 2757, 2730, 2700, 2670, 2654, 2621, 2594, 2580, 2548, 2516, 2499, 2485, 2460, 2426, 2414, 2392, 2373, 2356, 2330, 2312, 2293, 2272, 2259, 2235, 2220, 2207, 2183, 2171, 2152, 2129, 2117, 2104, 2085, 2077, 2055, 2038, 2033, 2007, 1999, 1989, 1971, 1953, 1943]"
"[3086, 3077, 3079, 3067, 3066, 3056, 3050, 3044, 3044, 3034, 3029, 3027, 3014, 3015, 3004, 3004, 2996, 2994, 2987, 2981, 2976, 2972, 2968, 2960, 2960, 2958, 2957, 2947, 2940, 2932, 2933, 2929, 2920, 2918, 2913, 2911, 2907, 2906, 2901, 2894, 2892, 2891, 2881, 2883, 2879, 2874, 2874, 2869, 2866, 2862, 2858, 2860, 2854, 2850, 2843, 2850, 2841, 2842, 2838, 2835, 2831, 2831, 2829, 2825, 2823, 2827, 2822, 2822, 2814, 2814, 2813, 2810, 2808, 2814, 2810, 2807, 2808, 2804, 2801, 2800, 2805, 2801, 2796, 2800, 2796, 2791, 2792, 2791, 2794, 2792, 2791, 2794, 2790, 2785, 2791, 2791, 2788, 2783, 2780, 2786, 2782, 2784, 2790, 2784, 2787, 2784, 2786, 2786, 2789, 2788, 2780, 2791, 2791, 2793, 2790, 2792, 2791, 2795, 2795, 2795, 2797, 2797, 2800, 2800, 2797, 2799, 2802, 2800, 2800, 2804, 2809, 2806, 2806, 2814, 2811, 2819, 2813, 2818, 2823, 2819, 2825, 2829, 2825, 2827, 2839, 2837, 2831, 2839, 2837, 2846, 2846, 2848, 2846, 2858, 2861, 2860, 2865, 2866, 2869, 2873, 2874, 2884, 2883, 2887, 2887, 2888, 2891, 2900, 2906, 2911, 2910, 2914, 2923, 2923, 2926, 2925, 2932, 2939, 2943, 2951, 2956, 2951, 2958, 2964, 2974, 2977, 2983, 2977, 2933, 2906, 2869, 2840, 2813, 2776, 2750, 2718, 2693, 2664, 2636, 2614, 2595, 2560, 2549, 2520, 2492, 2461, 2458, 2427, 2405, 2379, 2360, 2340, 2320, 2307, 2285, 2270, 2247, 2228, 2214, 2196, 2180, 2164, 2151, 2129, 2110, 2091, 2077, 2062, 2054, 2027, 2018, 2011, 1994, 1977, 1962, 1948, 1941]"

Here's is the continuous output, i am getting. Previously I was receiving data in Hex format. So I converted it in integer format. the range of my device is from 55 deg- 125 deg(i.e of 70 deg). It should plot a 3D graph where the x axis is from 0 to 5000mm the y axis from -500mm to 500mm. My sensor is fixed at one point. consider fixed point as origin. How do I go about this?

Edited Code as suggested in comment below

from __future__ import print_function
import socket
import sys
from time import sleep
import time
from mpl_toolkits.mplot3d import Axes3D
import csv
import matplotlib.pyplot as plt
import numpy as np
start = time.time()
def update_plot(data, a, ax, fig):
    ax.cla()  # removes previous plot
    # ax.plot(x, y, zs=0, zdir='z', label='curve in (x,y)')
    ax.scatter(np.cos(a) * data, np.sin(a) * data, zs=0, zdir='y',
               label='points in (x,z)')  # converst polar to XYZ for 3d
    ax.legend()
    ax.set_xlim(-1000, 1000)
    ax.set_ylim(-1000, 1000)
    ax.set_zlim(-200, 1000)
    ax.set_xlabel('X')
    ax.set_ylabel('Y')
    ax.set_zlabel('Z')
    ax.view_init(elev=20., azim=-35)
    plt.draw()
    plt.show()
    fig.canvas.flush_events()
    pass
def comm_dists(ip, port):
    sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
    server_address = (ip, port)
    print(sys.stderr, 'connecting to %s port %s' % server_address)
    sock.connect(server_address)
    print("Connected")
    try:
        message = b'\x02\x02\x02\x02\x00\x00\x00\x10sMI 0 3 F4724744 '
        sock.sendall(message)
        print(sys.stderr, 'sending "%s"' % message)
        print('sent "%s"' % message)
        data = sock.recv(5000)
        print(sys.stderr, 'received "%s"' % data)
        sleep(0.05)
        while True:
            data = sock.recv(5000)
            data = data.decode("utf-8", errors='ignore')
            data = data.split()
            data = [int.from_bytes(bytes.fromhex(item), 'big') for item in data[:-8]]
            fig = plt.figure()
            print('received "%s"' % data)
            ax = fig.gca(projection='3d')
            a = np.arange(11 * np.pi / 36, 24.65 * np.pi / 36, 0.05 * np.pi / 36)
            update_plot(data, a, ax, fig)
            time.sleep(0.05)
    finally:
        print(sys.stderr, 'closing socket')
i = 0
dados_csv = []

while i < int(1):
    data = comm_dists('192.168.0.1', 2111)
    i = i + 1

Edit Note: The sensor is fixed and object is moving on conveyor. Means object is passing through the light ray reflected from the sensor). As sensor position is considered at origin and is fixed. So the value of phi is constant. So, I considered 3rd parameter as zero. The output I am getting is r (distance between sensor and object). whereas, theta lies between 55 deg to 125 deg(70 deg). angle resolution is 0.25 deg. sliced sampled in no. of points in received values. eg. 1st angle = 55 deg. 2nd angle= 55.25...........nth angle=70 deg.

As Shown in the Figure

Edit Note 2:

However by putting the output values one by one in below code, I am able to get the correct 2D graph. I need help in getting 3D graph. Code:

import matplotlib.pyplot as plt
import numpy as np

lidar_bins = np.arange(55, 125, 0.25)
lidar_field = [[3088, 3081, 3078, 3068, 3066, 3058, 3051, 3046, 3036, 3032, 3028, 3018, 3018, 3006, 3007, 2999, 3001, 2994, 2986, 2984, 2977, 2974, 2967, 2960, 2960, 2954, 2947, 2944, 2940, 2931, 2930, 2925, 2920, 2918, 2915, 2909, 2907, 2904, 2899, 2896, 2891, 2885, 2881, 2884, 2876, 2874, 2874, 2862, 2866, 2862, 2860, 2854, 2851, 2846, 2851, 2844, 2839, 2841, 2835, 2841, 2829, 2827, 2823, 2831, 2829, 2815, 2821, 2825, 2821, 2816, 2810, 2815, 2812, 2807, 2810, 2807, 2810, 2805, 2802, 2796, 2797, 2798, 2802, 2795, 2792, 2792, 2795, 2790, 2787, 2794, 2787, 2793, 2789, 2785, 2787, 2788, 2790, 2782, 2782, 2789, 2786, 2780, 2784, 2790, 2783, 2788, 2781, 2784, 2789, 2791, 2788, 2787, 2793, 2793, 2790, 2793, 2793, 2793, 2793, 2795, 2791, 2791, 2799, 2800, 2795, 2800, 2806, 2802, 2799, 2807, 2809, 2815, 2814, 2811, 2817, 2808, 2817, 2816, 2827, 2825, 2820, 2831, 2836, 2834, 2834, 2838, 2835, 2840, 2854, 2851, 2847, 2854, 2856, 2857, 2857, 2863, 2863, 2867, 2873, 2873, 2882, 2877, 2882, 2891, 2887, 2895, 2902, 2897, 2908, 2911, 2913, 2920, 2919, 2929, 2929, 2931, 2939, 2941, 2947, 2956, 2961, 2962, 2967, 2971, 2978, 2981, 2978, 2935, 2906, 2858, 2838, 2798, 2775, 2744, 2713, 2692, 2670, 2645, 2615, 2580, 2560, 2539, 2502, 2494, 2473, 2447, 2428, 2407, 2387, 2371, 2348, 2326, 2298, 2285, 2259, 2247, 2227, 2218, 2187, 2173, 2164, 2138, 2125, 2106, 2098, 2082, 2071, 2047, 2031, 2018, 2003, 1990, 1974, 1960, 1950, 1939]]
ax = plt.subplot(111, projection='polar')
ax.set_rmax(5000)
ax.plot(np.deg2rad(lidar_bins), lidar_field)
plt.show()

I want my values to be fetched automatically and need to be plotted in 3D plot. How do do this? Output Graph

New Edit 3: I want to plot scatterplot from data received from the TCP port. As suggested in comment below, I have edited my above code accordingly. But It doesn't plot me the updated graph. and the output value I was receiving earlier as shown above stopped updating too.

shantmanu karna
  • 19
  • 3
  • How do you plan to plot a 3 dimensional plot with only 2 dimensions? You have point= (r,theta) and no third dimension. – Jason Chia Feb 28 '20 at 09:36
  • @Jason. The 3rd required value is Phi. The value of phi is constant. so I considered it as zero. The output I am getting is **r**. where as theta lies between 55 deg to 125 deg(**70 deg**). angle resolution is 0.25 deg. sliced sampled in no. of points in received values. eg. 1st angle = 55 deg. 2nd angle= 55.25...........nth angle=70 deg. – shantmanu karna Feb 28 '20 at 09:44
  • @JasonChia I have edited the question and explained my issue more clearly. I think it will help you in understanding it better. – shantmanu karna Feb 28 '20 at 09:56
  • I am too lazy to rewrite another answer as you deleted the on I made for you previously. So the solution is simple. Make a function to plot your data in whatever form you want by passing the data from your while loop. It should not recreate a figure or axes. Only update data. See the answer here for details.https://stackoverflow.com/questions/4098131/how-to-update-a-plot-in-matplotlib – Jason Chia Feb 28 '20 at 10:49
  • No @JasonChia , I didn't deleted any answer. This is my first question in this community. – shantmanu karna Feb 28 '20 at 11:26
  • @JasonChia Please help me in this. if you know how to do this. I would be very thankful. – shantmanu karna Feb 28 '20 at 11:28

1 Answers1

0

First you need to make sure that your received values are always the same length / all values are received. They should all be in the same dimension.

import time
import numpy as np
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

def test():
    plt.ion() #IMPORTANT
    fig = plt.figure()
    ax = fig.add_subplot(111,projection='3d')
    data = [[1,2,1],[2,1,7],[1,7,0]] #Now instead of using a for loop. simply call update plot when you have your received data
    theta = [np.pi,np.pi/2,0]#Define this in your original code
    for i in data:
        update_plot(data,theta,ax,fig)
        time.sleep(1)
def update_plot(data,theta,ax,fig):
    ax.cla()# removes previous plot
    ax.plot(data*np.cos(theta),data*np.sin(theta),0) #converst polar to XYZ for 3d
    plt.draw()
    fig.canvas.flush_events() #force plt to output the plot
Jason Chia
  • 1,144
  • 1
  • 5
  • 18
  • I have updated the code in my above question with your suggestion. – shantmanu karna Feb 29 '20 at 10:29
  • my received values is of length 273, whereas theta is between 55 deg. to 125 deg. with resolution of 0.25 deg. it make it length of data 280. So I am getting error **_ ax.plot(data*np.cos(theta),data*np.sin(theta),0) #converst polar to XYZ for 3d ValueError: operands could not be broadcast together with shapes (3,) (1,273) _** Is there a way I can resolve this error. – shantmanu karna Feb 29 '20 at 10:31
  • as I mentioned in the comments. You need to make sure that the length of your data = to length of your angles. Otherwise you are missing stuff. You can figure this out yourself. your 5 rows of values do not have the same length. So either add another step to put a dummy value for missing values. And this would be more of a systemic problem than python. You asked for a way to update 3d plot with R theta. I have given you that. your job to make sure len (r) == len(theta) – Jason Chia Mar 02 '20 at 09:28
  • I tried to create dummy values and also followed the suggestion as shown in the link. https://stackoverflow.com/questions/51404967/valueerror-operands-could-not-be-broadcast-together-with-shapes-2501-201-250 but No luck. It is still showing error: **`ValueError: operands could not be broadcast together with shapes (3,) (1,273)`** – shantmanu karna Mar 03 '20 at 08:44
  • One of your values have a dimension of 3. What is that? Please ensure that the shape of your arrays are correct. DO not use the theta from my code. You should generate your own range. theta should be the same shape as your input. – Jason Chia Mar 03 '20 at 10:18
  • When I am inserting my angle as the angle lies between 55 deg to 125 deg. of resolution 0.25 by using `lidar_bins = np.arange(55, 125, 0.25)` code. it is showing (3,) because (1,273) is the length of output i am receiving. – shantmanu karna Mar 03 '20 at 10:52
  • I tried a lot in achieving my output. But I am unable to achieve it. Can you please help me in getting it from scratch. I am adding my entire code if it help. Removing the new functions. It will be great help if you help in inserting the function with code. – shantmanu karna Mar 03 '20 at 10:57
  • Dimensions of lidar_bins must be equal to the array from integer_data. Can you verify that? Once that is done, simply call the function update_plot. You have to initialize the plot as in the function test I have given. Its not very difficult. – Jason Chia Mar 03 '20 at 11:56
  • I checked it. The dimension of integer data comming is 273 and lidar bins coming is 280. (70/0.25). So it was not equal. then I tried your suggestion of inserting dummy value as suggested above. It helped in making integer data dimension 280. i.e both value becomes of equal dimension. Then I used the above code to update_plot. When I run it. It is now showing me new error **`integer_data = [int.from_bytes(bytes.fromhex(item), 'big') for item in data[:-8]] ValueError: non-hexadecimal number found in fromhex() arg at position 5`** . – shantmanu karna Mar 03 '20 at 12:58
  • That is an entirely different problem. I would double check how you get your integer data and if the data is correct. Please ask another question for that problem. – Jason Chia Mar 03 '20 at 14:44
  • I got integer values from `integer_data = [int.from_bytes(bytes.fromhex(item), 'big') for item in data[:-8]]` using this code. It converts hex data received from TCP port into integer value. – shantmanu karna Mar 04 '20 at 04:42
  • Now, I am able to plot the 3D graph from my sensor. But, when I stop my program, an error is displayed in the end. – shantmanu karna Mar 04 '20 at 10:16