Merge pull request #3 from yuezhezhang/main

Adds m3p2i paper from Frankie

_data/publications.json

@@ -1,6 +1,29 @@
 [
-
-
+    {
+        "title": "Multi-Modal MPPI and Active Inference for Reactive Task and Motion Planning",
+        "authors": [
+            "Yuezhe Zhang",
+            "Corrado Pezzato",
+            "Elia Trevisan",
+            "Chadi Salmi",
+            "Carlos Hernández Corbato",
+            "Javier Alonso-Mora"
+        ],
+        "date": "2024-06-25",
+        "type": "journal",
+        "venue": "IEEE Robotics and Automation Letters (RA-L)",
+        "links": [
+            {
+                "pdf": "/assets/files/publications/24_zhang_ral_reactive_tamp.pdf",
+                "web": "/paper_websites/m3p2i-aip"
+            }
+        ],
+        "image": "/assets/images/papers/m3p2i_aip/rw_icon.png",
+        "belongs_to_projects": [
+            "trilogy", "airlab-manipulation"
+        ],
+        "abstract": "Task and Motion Planning (TAMP) has made strides in complex manipulation tasks, yet the execution robustness of the planned solutions remains overlooked. In this work, we propose a method for reactive TAMP to cope with runtime uncertainties and disturbances. We combine an Active Inference planner (AIP) for adaptive high-level action selection and a novel Multi-Modal Model Predictive Path Integral controller (M3P2I) for low-level control. This results in a scheme that simultaneously adapts both high-level actions and low-level motions. The AIP generates alternative symbolic plans, each linked to a cost function for M3P2I. The latter employs a physics simulator for diverse trajectory rollouts, deriving optimal control by weighing the different samples according to their cost. This idea enables blending different robot skills for fluid and reactive plan execution, accommodating plan adjustments at both the high and low levels to cope, for instance, with dynamic obstacles or disturbances that invalidate the current plan. We have tested our approach in simulations and real-world scenarios."
+      }, 
     {
         "title": "Simultaneous Synthesis and Verification of Neural Control Barrier Functions through Branch-and-Bound Verification-in-the-Loop Training",
         "authors": [

_msc_projects_finished/23_FrankieZhang_M3P2I-AIP.html

@@ -0,0 +1,40 @@
+---
+title: "Multi-Modal MPPI and Active Inference for Reactive Task and Motion Planning"
+authors:
+    - name: "Yuezhe Zhang"
+      url: "https://yuezhezhang.github.io/"
+      superscript: "1"
+    - name: "Corrado Pezzato"
+      url: "https://www.linkedin.com/in/corrado-pezzato/"
+      superscript: "1"
+    - name: "Elia Trevisan"
+      url: "https://www.linkedin.com/in/eliatrevisan/"
+      superscript: "1"
+    - name: "Chadi Salmi"
+      url: "https://c-salmi.github.io/"
+      superscript: "1"
+    - name: "Carlos Hernández Corbato"
+      url: "https://chcorbato.github.io/"
+      superscript: "1"
+    - name: "Javier Alonso-Mora"
+      url: "https://www.autonomousrobots.nl/"
+      superscript: "1"
+affiliations:
+    # - name: "Equal contribution"
+    #   superscript: "*"
+    - name: "TU Delft"
+      superscript: "1"
+      url: "https://tudelft.nl"
+end_date: 2023-09-01 # end date if ended, approximated if not sure. Just for display purposes and ordering.
+# This is the short project description, displayed in the project's card"
+description: "Task and Motion Planning (TAMP) has made strides in complex manipulation tasks, yet the execution robustness of the planned solutions remains overlooked. In this work, we
+propose a method for reactive TAMP to cope with runtime uncertainties and disturbances. We combine an Active Inference planner (AIP) for adaptive high-level action selection and a
+novel Multi-Modal Model Predictive Path Integral controller (M3P2I) for low-level control. This results in a scheme that simultaneously adapts both high-level actions and low-level motions.
+The AIP generates alternative symbolic plans, each linked to a cost function for M3P2I. The latter employs a physics simulator for diverse trajectory rollouts, deriving optimal control
+by weighing the different samples according to their cost. This idea enables blending different robot skills for fluid and reactive plan execution, accommodating plan adjustments at both the
+high and low levels to cope, for instance, with dynamic obstacles or disturbances that invalidate the current plan. We have tested our approach in simulations and real-world scenarios."
+cover_image: /assets/images/robots/AIRLab-1080x675.png # Image displayed in the project's card, make it aspect ratio 1x1 (square) for best results, and keep it a reasonable size (like 1-2MB). Can also be a gif
+---
+<head>
+  <meta http-equiv="refresh" content="0; url=../paper_websites/m3p2i-aip" />
+</head>