Skip to content

Commit 929533f

Browse files
SasiAlavalaSasiAlavala
committed
Update 6_4_26
1 parent 1a4f072 commit 929533f

4 files changed

Lines changed: 60 additions & 21 deletions

File tree

‎_pages/contact.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -17,7 +17,7 @@ permalink: /contact/
1717
<ul>
1818
<li>PhD applicants: Apply through the <a href="https://iisc.ac.in/admissions/" target="_blank">IISc admissions portal.</a></li>
1919
<li>Master's students: If you want to pursue your master's project, visit Prof. Ravi Prakash directly at his office.</li>
20-
<li>Project Associates and UG interns: Please email Prof. Ravi Prakash with your interests and CV.</li>
20+
<li>Project Associates and UG interns: Please email <a href="mailto:hiro.cps.iisc@gmail.com">hiro.cps.iisc@gmail.com</a> with your interests and CV.</li>
2121
</ul>
2222
</div>
2323

‎_pages/people.md‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -80,7 +80,7 @@ permalink: /people/
8080
<img src="/assets/img/people/sasidhar_alavala.JPG" alt="Sasidhar Alavala"/>
8181
</a>
8282
<div class="person-links">
83-
<a href="#" target="_blank" title="Personal Page"><i class="fas fa-globe"></i></a>
83+
<a href="https://sasidharalavala.work" target="_blank" title="Personal Page"><i class="fas fa-globe"></i></a>
8484
<a href="https://scholar.google.com/citations?user=1PStmj8AAAAJ&hl=en" target="_blank" title="Google Scholar"><i class="fas fa-graduation-cap"></i></a>
8585
<a href="https://www.linkedin.com/in/sasidharalavala/" target="_blank" title="LinkedIn"><i class="fab fa-linkedin"></i></a>
8686
</div>
@@ -113,7 +113,7 @@ permalink: /people/
113113
<a href="https://www.linkedin.com/in/akhil-r-kurup" target="_blank" title="LinkedIn"><i class="fab fa-linkedin"></i></a>
114114
</div>
115115
<span class="name">Akhil R Kurup</span>
116-
<span class="role">Error Detection and Recovery</span>
116+
<span class="role">MTech(R)</span>
117117
</div>
118118
<div class="person-card">
119119
<a href="https://mumukshtayal.github.io/" target="_blank" class="person-image-link">
@@ -137,7 +137,7 @@ permalink: /people/
137137
<a href="https://www.linkedin.com/in/ashrith9sagar/" target="_blank" title="LinkedIn"><i class="fab fa-linkedin"></i></a>
138138
</div>
139139
<span class="name">Ashrith Sagar Yedlapalli</span>
140-
<span class="role">Obstacle Avoidance</span>
140+
<span class="role">MTech</span>
141141
</div>
142142
</div>
143143

‎_pages/research.md‎

Lines changed: 56 additions & 17 deletions
Original file line numberDiff line numberDiff line change
@@ -25,41 +25,70 @@ permalink: /research/
2525
<div class="research-area-content">
2626
<h2>Interactive Reinforcement Learning</h2>
2727
<p>
28-
We develop methods for robots to learn complex tasks through interactive feedback from humans.
29-
Our work focuses on efficient learning from human corrections, incorporating human preferences,
30-
and enabling robots to actively query humans for guidance during learning and execution.
28+
We develop methods for robots to learn complex manipulation tasks through hierarchical reinforcement
29+
learning and imitation learning. Our work on <strong>Impedance Primitive-Augmented Hierarchical RL</strong>
30+
(ICRA 2025) augments high-level RL policies with impedance control primitives to solve long-horizon
31+
sequential tasks, enabling compliant and adaptive skill acquisition. We also investigate robust imitation
32+
learning — including methods that exploit <strong>mixed-quality demonstrations</strong> (Beyond the Teacher,
33+
ICRA 2026) and <strong>progress-aligned data curation</strong> (PACER, CoRL Workshop 2025) to improve
34+
generalization from imperfect human teachers, as well as stochastic encodings (RISE, IROS Workshop 2025)
35+
for robust policy learning.
3136
</p>
37+
<div class="pub-links-inline">
38+
<a href="https://ieeexplore.ieee.org/abstract/document/11128462" target="_blank" class="pub-link-text">ImpHRL (ICRA'25)</a>
39+
<a href="https://focaslab.github.io/beyondtheteacher/" target="_blank" class="pub-link-text">Beyond the Teacher (ICRA'26)</a>
40+
<a href="https://openreview.net/forum?id=gaYyBvP2Rz" target="_blank" class="pub-link-text">PACER</a>
41+
<a href="https://openreview.net/forum?id=GEexdUmA67" target="_blank" class="pub-link-text">RISE</a>
42+
</div>
3243
</div>
3344
<div class="research-area-image">
34-
<img src="/assets/img/research_areas/robot_learning.gif" alt="Interactive RL">
45+
<img src="/assets/img/publication_preview/icra_imphrl_25.gif" alt="Impedance Hierarchical RL">
3546
</div>
3647
</div>
3748

3849
<div class="research-area">
3950
<div class="research-area-content">
4051
<h2>Foundational Models for Robotics</h2>
4152
<p>
42-
We explore how large-scale pre-trained models can be leveraged for robotic manipulation and navigation.
43-
Our research includes vision-language models for task understanding, diffusion models for motion generation,
44-
and adapting foundation models to physical robot systems with limited data.
53+
We explore how large-scale pre-trained vision-language models can enable natural, flexible
54+
robot-human interaction. Our work on <strong>OVITA</strong> (Open-Vocabulary Interpretable Trajectory
55+
Adaptations, RA-L 2025) uses vision-language models to adapt robot trajectories in real time based on
56+
open-vocabulary human language instructions, providing interpretable modifications without retraining.
57+
We also develop <strong>DiffusionPack</strong> (NeurIPS Workshop 2025), a diffusion-based approach for
58+
complex bin-packing manipulation tasks that incorporates custom human preferences, bridging language,
59+
world models, and physical robot execution.
4560
</p>
61+
<div class="pub-links-inline">
62+
<a href="https://ieeexplore.ieee.org/abstract/document/11150730" target="_blank" class="pub-link-text">OVITA (RA-L'25)</a>
63+
<a href="https://anurag1000101.github.io/projects/IISC/" target="_blank" class="pub-link-text">Project Site</a>
64+
<a href="https://openreview.net/forum?id=uReNc199fG" target="_blank" class="pub-link-text">DiffusionPack</a>
65+
</div>
4666
</div>
4767
<div class="research-area-image">
48-
<img src="/assets/img/research_areas/language_robotics.gif" alt="Foundation Models">
68+
<img src="/assets/img/publication_preview/ovita_25.gif" alt="OVITA Trajectory Adaptation">
4969
</div>
5070
</div>
5171

5272
<div class="research-area">
5373
<div class="research-area-content">
5474
<h2>Safe and Compliant Human-Robot Interaction</h2>
5575
<p>
56-
Safety and compliance are essential when robots work alongside humans. We develop adaptive control strategies,
57-
variable impedance methods, and formal verification techniques to ensure safe physical interaction. Our research
58-
ensures robots can adapt their behavior based on human intent and environmental constraints.
76+
Safety and compliance are essential when robots work alongside humans. Our work on <strong>SafeDMPs</strong>
77+
(ICRA 2026) integrates Signal Temporal Logic (STL)-based formal safety specifications directly into Dynamic
78+
Movement Primitives, enabling adaptive robot motions that are provably safe during physical human-robot
79+
contact. We also develop <strong>certified reinforcement learning for variable impedance control</strong>
80+
(ICRA 2026), which combines Lyapunov-based stability certificates with RL to achieve optimal, safe impedance
81+
modulation during interaction. Our stability-aware PI² framework (ICRA Workshop 2025) further extends these
82+
ideas for safe interaction under uncertainty.
5983
</p>
84+
<div class="pub-links-inline">
85+
<a href="https://arxiv.org/abs/2509.16482" target="_blank" class="pub-link-text">SafeDMPs (ICRA'26)</a>
86+
<a href="https://tiwari-pranav.github.io/SafeDMPs/" target="_blank" class="pub-link-text">Project Site</a>
87+
<a href="https://openreview.net/forum?id=Xj3V96qTpf" target="_blank" class="pub-link-text">CoRL Workshop</a>
88+
</div>
6089
</div>
6190
<div class="research-area-image">
62-
<img src="/assets/img/research_areas/safe_hri.gif" alt="Safe HRI">
91+
<img src="/assets/img/publication_preview/safedmps_26.png" alt="Safe DMPs for HRI">
6392
</div>
6493
</div>
6594

@@ -74,21 +103,31 @@ permalink: /research/
74103
</p>
75104
</div>
76105
<div class="research-area-image">
77-
<img src="/assets/img/research_areas/optimal_control.gif" alt="3D Reconstruction">
106+
<img src="/assets/img/others/slam.png" alt="3D Reconstruction and SLAM">
78107
</div>
79108
</div>
80109

81110
<div class="research-area">
82111
<div class="research-area-content">
83112
<h2>Optimization and Optimal Control</h2>
84113
<p>
85-
We develop optimization-based approaches for robot motion planning and control. Our research
86-
includes adaptive optimal control for uncertain systems, trajectory optimization under constraints,
87-
and hierarchical planning frameworks that combine learning with classical optimization methods.
114+
We develop optimization-based and data-driven approaches for robot motion planning and control.
115+
Our <strong>Adaptive Critic</strong> framework (IEEE T-CST 2025) learns optimal controllers for uncertain
116+
robot manipulators using neural network-based value function approximation, achieving data-efficient online
117+
adaptation without explicit system identification. Our work on <strong>generalizable motion policies through
118+
keypoint parameterization and transportation maps</strong> (IEEE T-RO 2025) enables one-shot generalization
119+
of manipulation skills to novel object configurations. We also develop <strong>ST²</strong> (Sequentially
120+
Teaching Sequential Tasks, RA-M 2026), a framework for teaching robots long-horizon manipulation skills
121+
through sequential human demonstrations.
88122
</p>
123+
<div class="pub-links-inline">
124+
<a href="https://ieeexplore.ieee.org/abstract/document/10718695" target="_blank" class="pub-link-text">Adaptive Critic (T-CST'25)</a>
125+
<a href="https://ieeexplore.ieee.org/abstract/document/11049008" target="_blank" class="pub-link-text">Motion Policies (T-RO'25)</a>
126+
<a href="https://ieeexplore.ieee.org/document/11369949" target="_blank" class="pub-link-text">ST² (RA-M'26)</a>
127+
</div>
89128
</div>
90129
<div class="research-area-image">
91-
<img src="/assets/img/research_areas/optimal_control.gif" alt="Optimization">
130+
<img src="/assets/img/publication_preview/tcst_ac_24.gif" alt="Adaptive Critic Optimal Control">
92131
</div>
93132
</div>
94133

‎assets/img/others/slam.png‎

127 KB
Loading

0 commit comments

Comments
 (0)