From 26e641a477075984ff970a0338b9ba4a739d47fa Mon Sep 17 00:00:00 2001
From: dbrisbin <dbrisbin@umich.edu>
Date: Thu, 30 Apr 2020 19:42:26 -0400
Subject: [PATCH] planning functions notebook

---
 src/planning/Motion_Planning_Functions.ipynb | 611 +++++++++++++++++++
 1 file changed, 611 insertions(+)
 create mode 100644 src/planning/Motion_Planning_Functions.ipynb

diff --git a/src/planning/Motion_Planning_Functions.ipynb b/src/planning/Motion_Planning_Functions.ipynb
new file mode 100644
index 0000000..241a274
--- /dev/null
+++ b/src/planning/Motion_Planning_Functions.ipynb
@@ -0,0 +1,611 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "name": "Motion_Planning_Functions",
+      "provenance": [],
+      "collapsed_sections": []
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "accelerator": "GPU"
+  },
+  "cells": [
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "NoNMujanjNuu",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "#Given a map consisting of known poses and a start and end pose, find the optimal path between using A*\n",
+        "#Generate the relative motion in se2 between poses.\n",
+        "#This is straight line motion.\n",
+        "#Also implements cubic interpolation for a smooth trajectory across all points in path."
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "VqXKrUKwt05L",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "import matplotlib.pyplot as plt\n",
+        "import numpy as np\n",
+        "import random\n",
+        "import scipy.interpolate\n",
+        "import heapq #https://docs.python.org/3/library/heapq.html"
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "VMJWsOvIvGeJ",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "#Loading poses from the ground truth file\n",
+        "def load_poses(pose_gt_file) :\n",
+        "    pose_gt = np.loadtxt(pose_gt_file, delimiter = \",\")\n",
+        "    return pose_gt[1:, 1:3]\n",
+        "\n",
+        "poses = load_poses('../dataset/ground_truth/groundtruth_2012-01-08.csv')"
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "jQQn9T3U4yA2",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "#Astar and path functions\n",
+        "class PriorityQueue:\n",
+        "    def __init__(self):\n",
+        "        self.elements = []\n",
+        "    \n",
+        "    def empty(self):\n",
+        "        return len(self.elements) == 0\n",
+        "    \n",
+        "    def put(self, item, priority):\n",
+        "        heapq.heappush(self.elements, (priority, item))\n",
+        "    \n",
+        "    def get(self):\n",
+        "        return heapq.heappop(self.elements)[1]\n",
+        "\n",
+        "class Astar :\n",
+        "  # This class implements A* search along a network defined by several points\n",
+        "  # Poses is an array of coordinates\n",
+        "  # k defines how many nearest neighbors to look at during A* search\n",
+        "  # The primary usage of this class is the find_path function:\n",
+        "  #   Required parameters:\n",
+        "  #     start_idx:  \n",
+        "  #     goal_idx    \n",
+        "\n",
+        "  def __init__(self, poses) :\n",
+        "    self.poses = poses\n",
+        "    self.full_tree = scipy.spatial.KDTree(self.poses)\n",
+        "\n",
+        "  def _extract_path(self, cur_node, parent_idx, start_idx, sparse_poses):\n",
+        "    next_idx = cur_node\n",
+        "    path = [self.full_tree.query(sparse_poses[next_idx])[1]]\n",
+        "\n",
+        "    while next_idx != start_idx:\n",
+        "      next_idx = parent_idx[next_idx]\n",
+        "      path.append(self.full_tree.query(sparse_poses[next_idx])[1])\n",
+        "    \n",
+        "    return path[::-1]\n",
+        "\n",
+        "  def find_path(self, full_start_idx, full_goal_idx, sparseness=1, k=5):\n",
+        "    sparse_poses = poses[0::sparseness, :]\n",
+        "    \n",
+        "    visit_queue = PriorityQueue()\n",
+        "    visited_flag, queueed_flag = np.zeros(sparse_poses.shape[0]), np.zeros(sparse_poses.shape[0])\n",
+        "    g_score, h_score = np.full(sparse_poses.shape[0], np.inf), np.full(sparse_poses.shape[0], np.inf)\n",
+        "    parent_idx = np.zeros(sparse_poses.shape[0], dtype='int')\n",
+        "\n",
+        "    sparse_tree = scipy.spatial.KDTree(sparse_poses)\n",
+        "\n",
+        "    start_idx = sparse_tree.query(poses[full_start_idx])[1]\n",
+        "    goal_idx = sparse_tree.query(poses[full_goal_idx])[1]\n",
+        "   \n",
+        "    # initialize\n",
+        "    goal = sparse_poses[goal_idx]\n",
+        "\n",
+        "    g_score[start_idx] = 0\n",
+        "    visit_queue.put(start_idx, np.inf)\n",
+        "    queueed_flag[start_idx] = 1\n",
+        "    optimal = False\n",
+        "\n",
+        "    while not visit_queue.empty():\n",
+        "      cur_node = visit_queue.get()\n",
+        "      visited_flag[cur_node] = 1\n",
+        "\n",
+        "      if cur_node == goal_idx:\n",
+        "        optimal = True\n",
+        "        break\n",
+        "\n",
+        "      # find neighbours\n",
+        "      neighbors = sparse_tree.query(sparse_poses[cur_node], k=k)\n",
+        "\n",
+        "      for nb_cur_dist, nb_idx in zip(neighbors[0][1:], neighbors[1][1:]):\n",
+        "        if visited_flag[nb_idx] == 1:\n",
+        "          continue\n",
+        "\n",
+        "        temp_dist = g_score[cur_node] + np.linalg.norm(sparse_poses[cur_node] - sparse_poses[nb_idx])\n",
+        "        # temp_dist = g_score[cur_node] + nb_cur_dist     ## this not work\n",
+        "        if g_score[nb_idx] > temp_dist:\n",
+        "          g_score[nb_idx] = temp_dist\n",
+        "          parent_idx[nb_idx] = cur_node\n",
+        "          f_score = g_score[nb_idx] + np.linalg.norm(sparse_poses[nb_idx] - goal)\n",
+        "        \n",
+        "        # put into queen\n",
+        "        if queueed_flag[nb_idx] == 0:\n",
+        "          visit_queue.put(nb_idx, f_score)\n",
+        "          queueed_flag[nb_idx] = 1\n",
+        "\n",
+        "    path = self._extract_path(cur_node, parent_idx, start_idx, sparse_poses)    \n",
+        "    path[0] = full_start_idx\n",
+        "    path[-1] = full_goal_idx\n",
+        "\n",
+        "    return path, optimal\n",
+        "  \n",
+        "  def find_local_path(self, start_pose, path, steps=5) :\n",
+        "    set_trace()\n",
+        "\n",
+        "    path_tree = scipy.spatial.KDTree(self.poses[path])\n",
+        "    path_idx = path_tree.query(start_pose)[1]\n",
+        "    start_idx = self.full_tree.query(self.poses[path[path_idx]])[1]\n",
+        "\n",
+        "    if path_idx + 5 < len(path) :\n",
+        "      goal_idx =self.full_tree.query(self.poses[path[path_idx + steps]])[1]\n",
+        "    else :\n",
+        "      goal_idx =self.full_tree.query(self.poses[path[-1]])[1]\n",
+        "\n",
+        "    local_path, _ = self.find_path(start_idx, goal_idx)\n",
+        "\n",
+        "    return local_path\n",
+        "\n",
+        "def total_dist_fun(poses) :\n",
+        "  total_dist = 0\n",
+        "  curr_point = poses[0]\n",
+        "\n",
+        "  for idx in range(1, poses.shape[0]) :\n",
+        "    total_dist += np.linalg.norm(curr_point - poses[idx])\n",
+        "    curr_point = poses[idx]\n",
+        "  \n",
+        "  return total_dist \n"
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "cvOUf76S5zFZ",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "#construct A* instance\n",
+        "astar = Astar(poses)"
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "Mh_A8aKdvwEt",
+        "colab_type": "code",
+        "outputId": "5e84feb2-7cc6-4ab2-a4de-c873f3d7b96f",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 51
+        }
+      },
+      "source": [
+        "#Test A*\n",
+        "start_idx = np.random.randint(poses.shape[0])\n",
+        "goal_idx = np.random.randint(poses.shape[0])\n",
+        "\n",
+        "path, optimal = astar.find_path(start_idx, goal_idx, sparseness=10, k=50)"
+      ],
+      "execution_count": 0,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "8260\n",
+            "37563\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "CkqvtjJ65x4Y",
+        "colab_type": "code",
+        "outputId": "f538c889-e5dd-43a1-b615-213b9161cf75",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 635
+        }
+      },
+      "source": [
+        "#Plot computed path\n",
+        "plt.figure(figsize=(16,9))\n",
+        "plt.scatter(poses[:,1], poses[:,0], s=1)\n",
+        "plt.scatter(poses[path,1], poses[path,0], c='y', s=20)\n",
+        "plt.scatter(poses[start_idx,1], poses[start_idx,0], marker='o', c='g', s=500, label='start')\n",
+        "plt.scatter(poses[goal_idx,1], poses[goal_idx,0], marker='*', c='r', s=750, label='goal')\n",
+        "plt.legend()\n",
+        "plt.title('Ground Truth Position of Nodes with Overlaid A* Path')\n",
+        "plt.xlabel('East (m)')\n",
+        "plt.ylabel('North (m)')\n",
+        "plt.axis('equal')"
+      ],
+      "execution_count": 0,
+      "outputs": [
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "(-760.4307177277268,\n",
+              " 110.22309646812992,\n",
+              " -360.7839962824075,\n",
+              " 100.82784220736045)"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 318
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 1152x648 with 1 Axes>"
+            ]
+          },
+          "metadata": {
+            "tags": [],
+            "needs_background": "light"
+          }
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "CRtQebcnjeSM",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "#SE(2) functions\n",
+        "def matrix_log_SO2(SO2_mat) :\n",
+        "  #ln(R) in SO(3) = theta\n",
+        "  return np.arctan2(SO2_mat[1,0], SO2_mat[0, 0])\n",
+        "\n",
+        "def matrix_log_SE2(SE2_mat) :\n",
+        "  theta = matrix_log_SO2(SE2_mat[0:2, 0:2])\n",
+        "  if (theta < 1e-6) :\n",
+        "    A = 1\n",
+        "    B = 0\n",
+        "  else :\n",
+        "    A = np.sin(theta)/theta\n",
+        "    B = (1-np.cos(theta))/theta\n",
+        "\n",
+        "  v_inv = 1/(A**2 + B**2) * np.array([[A, B], [-B, A]])\n",
+        "  mat_log = np.array(np.matmul(v_inv, SE2_mat[0:2, 2]))\n",
+        "  mat_log = np.append(mat_log, theta)\n",
+        "\n",
+        "  return mat_log\n",
+        "\n",
+        "def matrix_exp_so2(theta) :\n",
+        "  #reconstruct R.\n",
+        "  return np.array([[np.cos(theta), -np.sin(theta)], [np.sin(theta), np.cos(theta)]])\n",
+        "\n",
+        "def matrix_exp_se2(twist) :\n",
+        "  theta = twist[-1]\n",
+        "  R = matrix_exp_so2(theta)\n",
+        "  \n",
+        "  #V converges to I2\n",
+        "  if (theta < 1e-6) :\n",
+        "    V = np.eye(2)\n",
+        "  else:\n",
+        "    V = 1/theta * np.array([[np.sin(theta), -(1 - np.cos(theta))], [(1-np.cos(theta)), np.sin(theta)]])\n",
+        "\n",
+        "  mat_exp = np.zeros((3,3))\n",
+        "  mat_exp[0:2, 0:2] = R\n",
+        "  mat_exp[0:2, 2] = np.matmul(V, twist[0:2])\n",
+        "  mat_exp[2, 2] = 1\n",
+        "  \n",
+        "  return mat_exp\n",
+        "\n",
+        "def get_twist_SE2(Xstart, pos_end, pos_future=None) :\n",
+        "  \n",
+        "  Xend = np.zeros((3,3))\n",
+        "  \n",
+        "  Xend[-1,-1] = 1\n",
+        "  Xend[0:2, 2] = pos_end\n",
+        "  \n",
+        "  #compute end direction (face in direction of future step i.e. end+1)\n",
+        "  if not pos_future is None:\n",
+        "    next_displacement = pos_future - pos_end\n",
+        "    next_theta = np.arctan2(next_displacement[1], next_displacement[0])\n",
+        "    Xend[0:2, 0:2] = np.array([[np.cos(next_theta), -np.sin(next_theta)], [np.sin(next_theta), np.cos(next_theta)]])\n",
+        "  else :\n",
+        "    Xend[0:2, 0:2] = Xstart[0:2, 0:2]\n",
+        "\n",
+        "  # set_trace()\n",
+        "  twist_SE2 = matrix_log_SE2(np.matmul(np.linalg.inv(Xstart), Xend))\n",
+        "\n",
+        "  return twist_SE2, Xend\n",
+        "\n",
+        "def twist_motion(Xstart, twist, s=1) :\n",
+        "  return np.matmul(Xstart, s * matrix_exp_se2(twist))"
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "XkTTjfR_EaGT",
+        "colab_type": "code",
+        "outputId": "5aeff481-c2f9-461b-e6e9-85e0f6b4ea12",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 136
+        }
+      },
+      "source": [
+        "print('testing exponential map for SE2')\n",
+        "thetas = [0, 1e-4, np.pi/2, np.pi, 15*np.pi/8, 4.5 * np.pi]\n",
+        "\n",
+        "for theta in thetas :\n",
+        "  test_SE2 = np.array([[np.cos(theta), -np.sin(theta), 1.5], [np.sin(theta), np.cos(theta), 2], [0, 0, 1]])\n",
+        "  \n",
+        "  twist = matrix_log_SE2(test_SE2)\n",
+        "  SE2_res = matrix_exp_se2(twist)\n",
+        "  \n",
+        "  assert(np.sum(test_SE2 - SE2_res) < 1e-6)\n",
+        "  print('passed theta = ', theta)"
+      ],
+      "execution_count": 0,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "testing exponential map for SE2\n",
+            "passed theta =  0\n",
+            "passed theta =  0.0001\n",
+            "passed theta =  1.5707963267948966\n",
+            "passed theta =  3.141592653589793\n",
+            "passed theta =  5.890486225480862\n",
+            "passed theta =  14.137166941154069\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "LsDb3F7fKBol",
+        "colab_type": "code",
+        "outputId": "fdeaf8f7-a326-4ea7-b495-7fb751b784e0",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 51
+        }
+      },
+      "source": [
+        "#Generation of twists and executing path\n",
+        "print('testing motion twist generation')\n",
+        "\n",
+        "pos_end = np.array([x1, y1])\n",
+        "pos_future = np.array([x2, y2])\n",
+        "\n",
+        "twists = []\n",
+        "\n",
+        "Xstart = np.eye(3)\n",
+        "Xstart[0:2, 2] = poses[path[0]]\n",
+        "\n",
+        "poses = np.array(poses)\n",
+        "\n",
+        "for pose_idx, path_idx in enumerate(path[1:-1]) :\n",
+        "  twist, Xstart = get_twist_SE2(Xstart, poses[pose_idx], poses[path_idx + 1])\n",
+        "  twists.append(twist)\n",
+        "  #print(twist)\n",
+        "\n",
+        "twist, Xend = get_twist_SE2(Xstart, poses[-1])\n",
+        "twists.append(twist)\n",
+        "\n",
+        "\n",
+        "Xk = np.eye(3)\n",
+        "Xk[0:2, 2] = poses[path[0]]\n",
+        "\n",
+        "for twist in twists :\n",
+        "  Xk = twist_motion(Xk, twist)\n",
+        "\n",
+        "assert(np.sum(Xk - Xend) < 1e-6)\n",
+        "print('passed')"
+      ],
+      "execution_count": 359,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "testing motion twist generation\n",
+            "passed\n"
+          ],
+          "name": "stdout"
+        }
+      ]
+    },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "PaYwXvSfeO4k",
+        "colab_type": "code",
+        "outputId": "e53f3ada-f8bd-4ec8-ffd2-124154658d74",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 985
+        }
+      },
+      "source": [
+        "#Cubic interpolation of poses.\n",
+        "\n",
+        "poss = np.array(poses[path])\n",
+        "velocities = np.zeros(poss.shape) \n",
+        "T = np.zeros(poss.shape[0])\n",
+        "\n",
+        "total_time = 100\n",
+        "total_dist = total_dist_fun(poss)\n",
+        "cum_dist = 0\n",
+        "\n",
+        "velocities[0] = 0\n",
+        "\n",
+        "for i in range(1, poss.shape[0] - 1) :\n",
+        "  seg_dist = np.linalg.norm(poss[i+1] - poss[i])\n",
+        "  velocities[i] = (((poss[i+1] - poss[i]) / seg_dist) + velocities[i-1])/2\n",
+        "  T[i] = total_time * cum_dist/total_dist\n",
+        "  cum_dist += seg_dist\n",
+        "  \n",
+        "T[-1] = total_time\n",
+        "velocities[-1] = 0\n",
+        "print(velocities)\n",
+        "\n",
+        "a = np.zeros((poss.shape[0], 4, poss.shape[1]))\n",
+        "\n",
+        "for j in range(0, poss.shape[0]-1) :\n",
+        "  del_Tj = T[j+1] - T[j]\n",
+        "  a[j, 0] = poss[j]\n",
+        "  a[j, 1] = velocities[j]\n",
+        "  a[j, 2] = (3 * poss[j+1] - 3 * poss[j] - 2 * velocities[j] * del_Tj - velocities[j+1] * del_Tj)/ (del_Tj**2)\n",
+        "  a[j, 3] = (2 * poss[j] + (velocities[j] + velocities[j+1]) * del_Tj - 2 * poss[j + 1]) / (del_Tj**3)\n",
+        "\n",
+        "del_t = 0.005\n",
+        "pos_x = [a[0,0][0]]\n",
+        "pos_y = [a[0,0][1]]\n",
+        "vel_x = [0]\n",
+        "vel_y = [0]\n",
+        "\n",
+        "total_trial = 100\n",
+        "\n",
+        "for t in np.arange(del_t, total_trial, del_t) :\n",
+        "  j = np.argmax(T > t)-1\n",
+        "  delta_t = t - T[j]\n",
+        "  pos_t = a[j, 0] + a[j, 1]* delta_t + a[j, 2] * (delta_t**2) + a[j, 3] * (delta_t**3)\n",
+        "  pos_x.append(pos_t[0])\n",
+        "  pos_y.append(pos_t[1])\n",
+        "  vel_x.append((pos_x[-1] - pos_x[-2])/del_t)\n",
+        "  vel_y.append((pos_y[-1] - pos_y[-2])/del_t)\n",
+        "\n",
+        "t = np.arange(0, total_trial, del_t)\n",
+        "\n",
+        "plt.figure(figsize=(16,9))\n",
+        "plt.plot(t[1:405], pos_x[1:405], linestyle='-', c='r', label='interpolated x position')\n",
+        "# plt.scatter(t[0:400], pos_y[0:400], label='y position')\n",
+        "\n",
+        "plt.scatter(T[1:14], poss[1:14,0], c='b')\n",
+        "plt.plot(T[1:14], poss[1:14,0], linestyle='-', c='g', label='no interpolation')\n",
+        "# plt.scatter(T[0:10], poss[0:10,1], label='y no interp')\n",
+        "\n",
+        "plt.legend()\n",
+        "plt.title('position with cubic interpolation of via points')\n",
+        "plt.xlabel('Time (s)')\n",
+        "plt.ylabel('Position (m)')\n",
+        "\n",
+        "\n",
+        "plt.figure()\n",
+        "plt.scatter(t[2:500], vel_x[2:500])\n",
+        "plt.scatter(t[2:500], vel_y[2:500])"
+      ],
+      "execution_count": 0,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "[[0.         0.        ]\n",
+            " [0.06225802 0.4961088 ]\n",
+            " [0.09777187 0.74359323]\n",
+            " ...\n",
+            " [0.61079301 0.79024087]\n",
+            " [0.64382541 0.76317733]\n",
+            " [0.         0.        ]]\n"
+          ],
+          "name": "stdout"
+        },
+        {
+          "output_type": "stream",
+          "text": [
+            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:27: RuntimeWarning: divide by zero encountered in true_divide\n",
+            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:28: RuntimeWarning: divide by zero encountered in true_divide\n"
+          ],
+          "name": "stderr"
+        },
+        {
+          "output_type": "execute_result",
+          "data": {
+            "text/plain": [
+              "<matplotlib.collections.PathCollection at 0x7efcb9cb42b0>"
+            ]
+          },
+          "metadata": {
+            "tags": []
+          },
+          "execution_count": 356
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 1152x648 with 1 Axes>"
+            ]
+          },
+          "metadata": {
+            "tags": [],
+            "needs_background": "light"
+          }
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "image/png": "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\n",
+            "text/plain": [
+              "<Figure size 432x288 with 1 Axes>"
+            ]
+          },
+          "metadata": {
+            "tags": [],
+            "needs_background": "light"
+          }
+        }
+      ]
+    }
+  ]
+}
\ No newline at end of file