-
Notifications
You must be signed in to change notification settings - Fork 0
/
publications.html
1107 lines (1004 loc) · 55.9 KB
/
publications.html
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
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=900">
<title>Publications</title>
<link rel="stylesheet" href="styles.css">
<link rel="icon" href="assets/icon.jpg">
<link href="https://fonts.googleapis.com/css?family=Google+Sans|Castoro"
rel="stylesheet">
<link href="https://fonts.googleapis.com/css?family=Noto+Sans:400,700,400italic,700italic"
rel="stylesheet">
<!-- <link href='http://fonts.googleapis.com/css?family=Lato:400,700,400italic,700italic' rel='stylesheet'> -->
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.15.3/css/all.min.css">
<style>
:root {
--highlight-color: rgba(255, 249, 196, 0.79)
}
</style>
</head>
<body>
<table width="900" border="0" align="center" cellspacing="0" cellpadding="20">
<tr>
<td style="width:25%; vertical-align:middle; padding-right: 10px;">
<a href="index.html"><img src="assets/logo.png" height="100"></a>
</td>
<td style="width:15%; vertical-align:middle; text-align:center; padding-right: 5px;">
<a href="people.html" style="font-size: 22px; color:black">People</a>
</td>
<td style="width:15%; vertical-align:middle; text-align:center; padding-right: 10px;">
<a href="publications.html" style="font-size: 22px; color:black">Publications</a>
</td>
<td style="width:15%; vertical-align:middle; text-align:center; padding-right: 0px;">
<a href="research.html" style="font-size: 22px; color:black">Research</a>
</td>
<td style="width:15%; vertical-align:middle; text-align:center; padding-right: 0px;">
<a href="robots.html" style="font-size: 22px; color:black">Robots</a>
</td>
<td style="width:15%; vertical-align:middle; text-align:center; padding-right: 0px;">
<a href="join.html" style="font-size: 22px; color:black">Join</a>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<p>
<!-- Check out the <a href="https://scholar.google.com/citations?user=joR1Z4UAAAAJ&hl=en&oi=ao" target="_blank">Google Scholar</a> page for a full and up-to-date publication list. -->
<sup>*</sup> denotes equal contribution and <sup>†</sup> denotes equal advising. Representative papers are <span style="background-color: var(--highlight-color)">highlighted</span>.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<heading>Preprints</heading>
</td>
</tr>
</table>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_HOVER.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots</papertitle>
<br>
Tairan He<sup>*</sup>, Wenli Xiao<sup>*</sup>, Toru Lin, Zhengyi Luo, Zhenjia Xu, Zhenyu Jiang, Jan Kautz, Changliu Liu, Guanya Shi, Xiaolong Wang, Linxi Fan<sup>†</sup>, Yuke Zhu<sup>†</sup>
<br>
<a href="https://arxiv.org/abs/2410.21229" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://hover-versatile-humanoid.github.io/" target="_blank"><i class="fas fa-globe"></i> website</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: HOVER unifies many humanoid whole-body control modes to one policy that supports diverse control modes and outperforms each specialist.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_MPPI_uncertainty.png' width="95%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Agile Mobility with Rapid Online Adaptation via Meta-Learning and Uncertainty-Aware MPPI</papertitle>
<br>
Dvij Kalaria, Haoru Xue, Wenli Xiao, Tony Tao, Guanya Shi, John Dolan
<br>
<a href="https://arxiv.org/abs/2410.06565" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/meta-learning-model-adaptation" target="_blank"><i class="fas fa-globe"></i> website</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Meta-learn a vehicle dynamics model ensemble and use uncertainty-aware MPPI for control.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_SSML-AC.png' width="62%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Self-Supervised Meta-Learning for All-Layer DNN-Based Adaptive Control with Stability Guarantees</papertitle>
<br>
Guanqi He, Yogita Choudhary, Guanya Shi
<br>
<a href="https://arxiv.org/abs/2410.07575" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/ssml-ac-project" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/SSML-AC/tree/main" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Pretrain a residual dynamics DNN using meta-learning and fine-tune the whole DNN online using adaptive control with stability guarantees.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_AnyCar.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>AnyCar to Anywhere: Learning Universal Dynamics Model for Agile and Adaptive Mobility</papertitle>
<br>
Wenli Xiao<sup>*</sup>, Haoru Xue<sup>*</sup>, Tony Tao, Dvij Kalaria, John M. Dolan, Guanya Shi
<br>
<a href="https://arxiv.org/abs/2409.15783" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://lecar-lab.github.io/anycar/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/anycar" target="_blank"><i class="fas fa-code"></i> code</a>  
<a href="https://spectrum.ieee.org/video-friday-mobile-robot-upgrades" target="_blank"><i class="fas fa-newspaper"></i> IEEE Spectrum</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: AnyCar is a transformer-based dynamics model that can adapt to various vehicles, environments, state estimators, and tasks.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_DIAL-MPC.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealing</papertitle>
<br>
Haoru Xue<sup>*</sup>, Chaoyi Pan<sup>*</sup>, Zeji Yi, Guannan Qu, Guanya Shi
<br>
<a href="https://arxiv.org/abs/2409.15610" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://lecar-lab.github.io/dial-mpc/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/dial-mpc" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: DIAL-MPC is the first training-free method achieving real-time whole-body torque control using full-order dynamics.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_JumpingCoD.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Agile Continuous Jumping in Discontinuous Terrains</papertitle>
<br>
Yuxiang Yang, Guanya Shi, Changyi Lin, Xiangyun Meng, Rosario Scalise, Mateo Guaman Castro, Wenhao Yu, Tingnan Zhang, Ding Zhao, Jie Tan, Byron Boots
<br>
<a href="https://arxiv.org/abs/2409.10923" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://yxyang.github.io/jumping_cod/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/yxyang/jumping_cod" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Continuous, agile, and autonomous quadrupedal jumping via hierarchical model-free RL and model-based control.
</p>
</td>
</tr>
</table>
<br>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<heading>2024</heading>
</td>
</tr>
</table>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_MBD.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Model-Based Diffusion for Trajectory Optimization</papertitle>
<br>
Chaoyi Pan<sup>*</sup>, Zeji Yi<sup>*</sup>, Guanya Shi<sup>†</sup>, Guannan Qu<sup>†</sup>
<br>
<em>Neural Information Processing Systems (NeurIPS)</em>, 2024
<br>
<a href="https://drive.google.com/file/d/1kPjD79Cfr9spWulWNVFMRHqTE-mjbGAp/view?usp=sharing" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://lecar-lab.github.io/mbd/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/model-based-diffusion" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: MBD is a diffusion-based traj optimization method that directly computes the score function using models without any external data.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_flying_calligrapher.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Flying Calligrapher: Contact-Aware Motion and Force Planning and Control for Aerial Manipulation</papertitle>
<br>
Xiaofeng Guo<sup>*</sup>, Guanqi He<sup>*</sup>, Jiahe Xu, Mohammadreza Mousaei, Junyi Geng, Sebastian Scherer, Guanya Shi
<br>
<em>IEEE Robotics and Automation Letters (RA-L)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2407.05587" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://xiaofeng-guo.github.io/flying-calligrapher/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://spectrum.ieee.org/video-friday-unitree-talks-robots" target="_blank"><i class="fas fa-newspaper"></i> IEEE Spectrum</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Flying calligrapher enables precise hybrid motion and contact force control for an aerial manipulator in various drawing tasks.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='assets/OmniH2O.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning</papertitle>
<br>
Tairan He<sup>*</sup>, Zhengyi Luo<sup>*</sup>, Xialin He<sup>*</sup>, Wenli Xiao, Chong Zhang, Weinan Zhang, Kris Kitani, Changliu Liu, Guanya Shi
<br>
<em>Conference on Robot Learning (CoRL)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2406.08858" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://omni.human2humanoid.com/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://cmu.box.com/s/kmayzq5ax2rxvwn97s0hzz0aq5vws9io" target="_blank"><i class="fas fa-database"></i> dataset</a>  
<a href="https://github.com/LeCAR-Lab/human2humanoid" target="_blank"><i class="fas fa-code"></i> code</a>  
<a href="https://spectrum.ieee.org/video-friday-drone-vs-flying-canoe" target="_blank"><i class="fas fa-newspaper"></i> IEEE Spectrum</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: OmniH2O provides a universal whole-body humanoid control interface that enables diverse teleoperation and autonomy methods.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='assets/WoCoCo.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>WoCoCo: Learning Whole-Body Humanoid Control with Sequential Contacts</papertitle>
<br>
Chong Zhang<sup>*</sup>, Wenli Xiao<sup>*</sup>, Tairan He, Guanya Shi
<br>
<em>Conference on Robot Learning (CoRL)</em>, 2024
<p style="color: orange; margin: 0px 0;">(Oral Presentation)</p>
<a href="https://arxiv.org/abs/2406.06005" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://lecar-lab.github.io/wococo/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://spectrum.ieee.org/video-friday-drone-vs-flying-canoe" target="_blank"><i class="fas fa-newspaper"></i> IEEE Spectrum</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: WoCoCo is a task-agnostic skill learning framework without any motion priors, by decomposing long-horizon tasks into contact sequences.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='assets/H2O.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Learning Human-to-Humanoid Real-Time Whole-Body Teleoperation</papertitle>
<br>
Tairan He<sup>*</sup>, Zhengyi Luo<sup>*</sup>, Wenli Xiao, Chong Zhang, Kris Kitani, Changliu Liu, Guanya Shi
<br>
<em>International Conference on Intelligent Robots and Systems (IROS)</em>, 2024
<p style="color: orange; margin: 0px 0;">(Oral Presentation)</p>
<a href="https://arxiv.org/abs/2403.04436" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://human2humanoid.com/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/human2humanoid" target="_blank"><i class="fas fa-code"></i> code</a>  
<a href="https://spectrum.ieee.org/video-friday-human-to-humanoid" target="_blank"><i class="fas fa-newspaper"></i> IEEE Spectrum</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: H2O enables real-time whole-body teleoperation of a full-sized humanoid to perform tasks like pick and place, walking, kicking, boxing, etc.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='assets/abs.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion</papertitle>
<br>
Tairan He<sup>*</sup>, Chong Zhang<sup>*</sup>, Wenli Xiao, Guanqi He, Changliu Liu, Guanya Shi
<br>
<em>Robotics: Science and Systems (RSS)</em>, 2024
<p style="color: orange; margin: 0px 0;">(Outstanding Student Paper Award Finalist)</p>
<a href="https://arxiv.org/abs/2401.17583" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://agile-but-safe.github.io/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/ABS" target="_blank"><i class="fas fa-code"></i> code</a>  
<a href="https://spectrum.ieee.org/video-friday-agile-but-safe" target="_blank"><i class="fas fa-newspaper"></i> IEEE Spectrum</a>  
<a href="https://www.ri.cmu.edu/collision-free-high-speed-robots/" target="_blank"><i class="fas fa-newspaper"></i> CMU News</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: ABS enables fully onboard, agile (>3m/s), and collision-free locomotion for quadrupedal robots in cluttered environments.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_offroad.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Model Predictive Control for Aggressive Driving Over Uneven Terrain</papertitle>
<br>
Tyler Han, Alex Liu, Anqi Li, Alex Spitzer, Guanya Shi, Byron Boots
<br>
<em>Robotics: Science and Systems (RSS)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2311.12284" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/cs.washington.edu/off-road-mpc" target="_blank"><i class="fas fa-globe"></i> website</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We design a terrain-aware MPC framework that enables agile driving over uneven offroad geometries such as hills, banks, and ditches.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_DPO-IK.png' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Propagative Distance Optimization for Constrained Inverse Kinematics</papertitle>
<br>
Yu Chen, Yilin Cai, Jinyun Xu, Zhongqiang Ren, Guanya Shi, Howie Choset
<br>
<em>International Workshop on the Algorithmic Foundations of Robotics (WAFR)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2406.11572" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We propose a fast and scalable method for high-dim constrained IK problems, based on propagative distance-based optimization.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_CoVO-MPC.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>CoVO-MPC: Theoretical Analysis of Sampling-based MPC and Optimal Covariance Design</papertitle>
<br>
Zeji Yi<sup>*</sup>, Chaoyi Pan<sup>*</sup>, Guanqi He, Guannan Qu<sup>†</sup>, Guanya Shi<sup>†</sup>
<br>
<em>Conference on Learning for Dynamics and Control (L4DC)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2401.07369" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://lecar-lab.github.io/CoVO-MPC/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/LeCAR-Lab/CoVO-MPC" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We quantify the convergence rate of sampling-based MPC, and design a practical and effective algorithm CoVO-MPC with optimal rate.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_HMAC.png ' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Hierarchical Meta-learning-based Adaptive Controller</papertitle>
<br>
Fengze Xie, Guanya Shi, Michael O'Connell, Yisong Yue, Soon-Jo Chung
<br>
<em>International Conference on Robotics and Automation (ICRA)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2311.12367" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/hmacproject" target="_blank"><i class="fas fa-globe"></i> website</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: HMAC handles both manageable and latent disturbances with hierarchical iterative learning and smoothed streaming meta-learning.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_SafeDPA.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Safe Deep Policy Adaptation</papertitle>
<br>
Wenli Xiao<sup>*</sup>, Tairan He<sup>*</sup>, John Dolan, Guanya Shi
<br>
<em>International Conference on Robotics and Automation (ICRA)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2310.08602" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/safe-deep-policy-adaptation" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://youtu.be/PkyRzlRQVbE?si=B3guhmFEJyFhhgiS" target="_blank"><i class="fas fa-video"></i> video</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: SafeDPA jointly tackles the problems of policy adaptation and safe reinforcement learning, under unseen disturbances in the real world.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_DMPO.png' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Deep Model Predictive Optimization</papertitle>
<br>
Jacob Sacks, Rwik Rana, Kevin Huang, Alex Spitzer, Guanya Shi, Byron Boots
<br>
<em>International Conference on Robotics and Automation (ICRA)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2310.04590" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/uw.edu/dmpo" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/jisacks/dmpo" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: DMPO learns the inner-loop optimizer of sampling-based MPC directly via experience, outperforming MPC and end-to-end RL baselines.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_aerial_interaction.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Aerial Interaction with Tactile Sensing</papertitle>
<br>
Xiaofeng Guo, Guanqi He, Mohammadreza Mousaei, Junyi Geng, Guanya Shi, Sebastian Scherer
<br>
<em>International Conference on Robotics and Automation (ICRA)</em>, 2024
<br>
<a href="https://arxiv.org/abs/2310.00142" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/aerial-system-gelsight" target="_blank"><i class="fas fa-globe"></i> website</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We introduce a new aerial manipulation system that leverages tactile feedback for accurate contact force control and texture detection.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2024_gyf.gif' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Guardians as You Fall: Active Mode Transition for Safe Falling</papertitle>
<br>
Yikai Wang, Mengdi Xu, Guanya Shi, Ding Zhao
<br>
<em>IEEE International Automated Vehicle Validation Conference (IAVVC)</em>, 2024
<p style="color: orange; margin: 0px 0;">(Best Paper Award - Innovation)</p>
<a href="https://arxiv.org/abs/2310.04828" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/guardians-as-you-fall/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/ykwang20/Guardians_as_You_Fall" target="_blank"><i class="fas fa-code"></i> code</a>  
<a href="https://youtu.be/e0ORrUjrncc?si=oQwIA9tf8VHGk8D6" target="_blank"><i class="fas fa-video"></i> video</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: GYF is a safe falling and recovery framework that can actively tumble and recover to stable modes to reduce damage.
</p>
</td>
</tr>
</table>
<br>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<heading>2023</heading>
</td>
</tr>
</table>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2023_optimal_exploration.png' width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Optimal Exploration for Model-based RL in Nonlinear Systems</papertitle>
<br>
Andrew Wagenmaker, Guanya Shi, Kevin Jamieson
<br>
<em>Neural Information Processing Systems (NeurIPS)</em>, 2023
<p style="color: orange; margin: 0px 0;">(Spotlight, 3.1%)</p>
<a href="https://arxiv.org/abs/2306.09210" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Not all model parameters are equally important. We develop an instance-optimal exploration algorithm for MBRL in nonlinear systems.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2023_active_learning.png' width="85%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Active Representation Learning for General Task Space with Applications in Robotics</papertitle>
<br>
Yifang Chen, Yingbing Huang, Simon S. Du, Kevin Jamieson, Guanya Shi
<br>
<em>Neural Information Processing Systems (NeurIPS)</em>, 2023
<br>
<a href="https://arxiv.org/abs/2306.08942" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Inspired by robotics applications, we study algorithms for active representation learning with continuous task parametrization.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2023_datt.gif' width="62%">
<img src='publications/2023_datt.png' width="38%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>DATT: Deep Adaptive Trajectory Tracking for Quadrotor Control</papertitle>
<br>
Kevin Huang, Rwik Rana, Alexander Spitzer, Guanya Shi, Byron Boots
<br>
<em>Conference on Robot Learning (CoRL)</em>, 2023
<p style="color: orange; margin: 0px 0;">(Oral presentation, 6.6%)</p>
<a href="https://openreview.net/pdf?id=XEw-cnNsr6" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/deep-adaptive-traj-tracking" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/KevinHuang8/DATT" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: DATT can precisely track arbitrary, potentially infeasible trajectories in the presence of large disturbances.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src='publications/2023_cajun.gif' width="100%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>CAJun: Continuous Adaptive Jumping using a Learned Centroidal Controller</papertitle>
<br>
Yuxiang Yang, Guanya Shi, Xiangyun Meng, Wenhao Yu, Tingnan Zhang, Jie Tan, Byron Boots
<br>
<em>Conference on Robot Learning (CoRL)</em>, 2023
<br>
<a href="https://arxiv.org/abs/2306.09557" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://youtu.be/uWofjeirdEY" target="_blank"><i class="fas fa-video"></i> video</a>  
<a href="https://yxyang.github.io/cajun/" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://github.com/yxyang/cajun" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: CAJun is a hierarchical learning and control framework that enables legged robots to jump continuously with adaptive distances.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<img src='publications/2023_power_adaptive.png' width="100%">
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Leveraging Predictions in Power System Frequency Control: an Adaptive Approach</papertitle>
<br>
Wenqi Cui, Guanya Shi, Yuanyuan Shi, Baosen Zhang
<br>
<em>IEEE Conference on Decision and Control (CDC)</em>, 2023
<br>
<a href="https://arxiv.org/abs/2305.12044" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We combine adaptive nonlinear control and neural control for frequency restoration in power systems.
</p>
</td>
</tr>
</table>
<br>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<heading>2022</heading>
</td>
</tr>
</table>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<img src='publications/2022_neural_fly.gif' width="90%">
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Neural-Fly Enables Rapid Learning for Agile Flight in Strong Winds</papertitle>
<br>
Michael O'Connell<sup>*</sup>, Guanya Shi<sup>*</sup>, Xichen Shi, Kamyar Azizzadenesheli, Animashree Anandkumar, Yisong Yue, Soon-Jo Chung
<br>
<em>Science Robotics</em>
<br>
<a href="https://arxiv.org/abs/2205.06908" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://youtu.be/TuF9teCZX0U" target="_blank"><i class="fas fa-video"></i> video</a>  
<a href="https://www.caltech.edu/about/news/rapid-adaptation-of-deep-learning-teaches-drones-to-survive-any-weather" target="_blank"><i class="fas fa-newspaper"></i> Caltech news</a>  
<a href="https://youtu.be/R1S5BnKgJxs" target="_blank"><i class="fas fa-newspaper"></i> Reuters</a>  
<a href="https://www.cnn.com/videos/business/2022/05/31/caltech-neural-fly-drones-in-strong-wind-orig-ht.cnn-business/video/playlists/business-tech/" target="_blank"><i class="fas fa-newspaper"></i> CNN</a>  
<a href="https://github.com/aerorobotics/neural-fly" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Neural-Fly uses adaptive control to online fine-tune a meta-pretrained DNN representation, enabling rapid adaptation in strong winds.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2022_soco_delay_nonlinear.png" width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Online Optimization with Feedback Delay and Nonlinear Switching Cost</papertitle>
<br>
Weici Pan, Guanya Shi, Yiheng Lin, Adam Wierman
<br>
<em>Proceedings of the ACM on Measurement and Analysis of Computing Systems (SIGMETRICS)</em>
<br>
<a href="https://dl.acm.org/doi/abs/10.1145/3508037" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We propose a new online optimization setting with delay and nonlinear switching cost, and provide compeititve algorithms.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2022_robustness_consistency.png" width="100%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Robustness and Consistency in Linear Quadratic Control with Predictions</papertitle>
<br>
Tongxin Li<sup>*</sup>, Ruixiao Yang<sup>*</sup>, Guannan Qu, Guanya Shi, Chenkai Yu, Adam Wierman, Steven Low
<br>
<em>Proceedings of the ACM on Measurement and Analysis of Computing Systems (SIGMETRICS)</em>
<br>
<a href="https://dl.acm.org/doi/abs/10.1145/3508038" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: For online control with noisy predictions, we design an algorithm to optimally balance robustness and consistency (performance if no noise).
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2022_delayed_imperfect_information.png" width="100%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Competitive Control with Delayed Imperfect Information</papertitle>
<br>
Chenkai Yu, Guanya Shi, Soon-Jo Chung, Yisong Yue, Adam Wierman
<br>
<em>American Control Conference (ACC)</em>, 2022
<br>
<a href="https://ieeexplore.ieee.org/abstract/document/9867421" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://www.gshi.me/blog/CompetitiveMPC/" target="_blank"><i class="fas fa-file-alt"></i> blog</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We study MPC's dynamic regret and competitive ratio in the presence of prediction inaccuracy and feedback delay.
</p>
</td>
</tr>
</table>
<br>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<heading>2021</heading>
</td>
</tr>
</table>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2021_perturbation_mpc.png" width="100%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Perturbation-based Regret Analysis of Predictive Control in LTV Systems</papertitle>
<br>
Yiheng Lin<sup>*</sup>, Yang Hu<sup>*</sup>, Guanya Shi<sup>*</sup>, Haoyuan Sun<sup>*</sup>, Guannan Qu<sup>*</sup>, Adam Wierman
<br>
<em>Neural Information Processing Systems (NeurIPS)</em>, 2021
<p style="color: orange; margin: 0px 0;">(Spotlight, 2.9%)</p>
<a href="https://proceedings.neurips.cc/paper/2021/hash/298f587406c914fad5373bb689300433-Abstract.html" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://www.gshi.me/blog/CompetitiveMPC/" target="_blank"><i class="fas fa-file-alt"></i> blog</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We prove MPC's dynamic regret and competitive ratio exponentially improve as its prediction gets longer, in LTV systems.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2021_OMAC.gif" width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Meta-Adaptive Nonlinear Control: Theory and Algorithms</papertitle>
<br>
Guanya Shi, Kamyar Azizzadenesheli, Michael O'Connell, Soon-Jo Chung, Yisong Yue
<br>
<em>Neural Information Processing Systems (NeurIPS)</em>, 2021
<br>
<a href="https://arxiv.org/abs/2106.06098" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://github.com/GuanyaShi/Online-Meta-Adaptive-Control" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We present an online multi-task learning approach for adaptive nonlinear control with non-asymptotic guarantees.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2021_fast_uq.png" width="85%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Fast Uncertainty Quantification for Deep Object Pose Estimation</papertitle>
<br>
Guanya Shi, Yifeng Zhu, Jonathan Tremblay, Stan Birchfield, Fabio Ramos, Animashree Anandkumar, Yuke Zhu
<br>
<em>International Conference on Robotics and Automation (ICRA)</em>, 2021
<br>
<a href="https://arxiv.org/abs/2011.07748" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://sites.google.com/view/fastuq" target="_blank"><i class="fas fa-globe"></i> website</a>  
<a href="https://developer.nvidia.com/blog/nvidia-research-fast-uncertainty-quantification-for-deep-object-pose-estimation/" target="_blank"><i class="fas fa-file-alt"></i> blog</a>  
<a href="https://github.com/NVlabs/DOPE-Uncertainty" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We develop a simple and efficient UQ method for 6-DoF pose estimation, and apply it in real-world grasping tasks.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<img src='publications/2021_neural_swarm.gif' width="90%">
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Neural-Swarm2: Planning and Control of Heterogeneous Multirotor Swarms Using Learned Interactions</papertitle>
<br>
Guanya Shi, Wolfgang Hönig, Xichen Shi, Yisong Yue, Soon-Jo Chung
<br>
<em>IEEE Transactions on Robotics</em>
<br>
<a href="https://arxiv.org/abs/2012.05457" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://youtu.be/Y02juH6BDxo" target="_blank"><i class="fas fa-video"></i> video</a>  
<a href="https://www.caltech.edu/about/news/machine-learning-helps-robot-swarms-coordinate" target="_blank"><i class="fas fa-newspaper"></i> Caltech news</a>  
<a href="https://news.yahoo.com/caltech-drone-swarm-ai-174642584.html" target="_blank"><i class="fas fa-newspaper"></i> Yahoo! news</a>  
<a href="https://github.com/aerorobotics/neural-swarm" target="_blank"><i class="fas fa-code"></i> code</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: Neural-Swarm is a learning-based controller and planner for close-proximity flight of heterogeneous multirotor swarms.
</p>
</td>
</tr>
</table>
<br>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td width="100%" valign="middle">
<heading>2020</heading>
</td>
</tr>
</table>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2020_info_snoc.png" width="80%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Chance-Constrained Trajectory Optimization for Safe Exploration and Learning of Nonlinear Systems</papertitle>
<br>
Yashwanth Kumar Nakka, Anqi Liu, Guanya Shi, Animashree Anandkumar, Yisong Yue, Soon-Jo Chung
<br>
<em>IEEE Robotics and Automation Letters (RA-L)</em>
<br>
<a href="https://arxiv.org/abs/2005.04374" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://yashwanthnakka.com/trajectory-optimization-for-safe-exploration/" target="_blank"><i class="fas fa-file-alt"></i> blog</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We derive an iterative algorithm to solve information-cost stochastic nonlinear optimal control problems for safe episodic learning.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<img src='publications/2020_robust_regression.png' width="100%">
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Robust Regression for Safe Exploration in Control</papertitle>
<br>
Anqi Liu, Guanya Shi, Soon-Jo Chung, Animashree Anandkumar, Yisong Yue
<br>
<em>Conference on Learning for Dynamics and Control (L4DC)</em>, 2020
<br>
<a href="https://arxiv.org/pdf/1906.05819" target="_blank"><i class="far fa-file"></i> paper</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We derive generalization bounds under domain shift and connect them with safety bounds in control, for end-to-end safe explorations.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr style="background-color: var(--highlight-color)">
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<img src='publications/2020_mpc.png' width="100%">
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>The Power of Predictions in Online Control</papertitle>
<br>
Chenkai Yu, Guanya Shi, Soon-Jo Chung, Yisong Yue, Adam Wierman
<br>
<em>Neural Information Processing Systems (NeurIPS)</em>, 2020
<br>
<a href="https://arxiv.org/abs/2006.07569" target="_blank"><i class="far fa-file"></i> paper</a>  
<a href="https://videos.neurips.cc/search/power%20of%20predictions%20in%20online%20control/video/slideslive-38936069" target="_blank"><i class="fas fa-video"></i> video</a>  
<a href="https://www.gshi.me/blog/CompetitiveMPC/" target="_blank"><i class="fas fa-file-alt"></i> blog</a>
<p style="margin-top: 5px"><i class="fas fa-comment-dots"></i> TL;DR: We give the first non-asymptotic guarantee for MPC. MPC's dynamic regret exponentially decreases as its prediction gets longer.
</p>
</td>
</tr>
</table>
<br>
<table width="880" border="0" align="center" cellspacing="0" cellpadding="0">
<tr>
<td style="width:35%; vertical-align:middle; padding-right: 20px;">
<div class="image-container">
<img src="publications/2020_optimistic_robd.png" width="90%">
</div>
</td>
<td style="width:65%; vertical-align:middle">
<papertitle>Online Optimization with Memory and Competitive Control</papertitle>