figure 1. a sampling of representative work in evolutionary robotics.

evolutionary robotics often involves optimizing not only the controller of a robot but also its body plan. Formal grammars (a12) and algorithms that simulate development (b1) have been used to optimize robots in simulation.

Additive manufacturing has been employed to build physical versions of evolved simulated robots semiautomatically (c,d18). Once deployed as physical machines, evolutionary algorithms have been used to allow damaged robots to recover from injury (e4) as well as ease the transferral of newly evolved controllers from simulation to the physical robot (f16).

In addition to locomotion, researchers have evolved more cognitively demanding behaviors such as discriminating between differently shaped objects by manipulating them (g35) or physically demanding tasks like aerial swarming (h11). Behaviors have also been evolved for robots with non-traditional body plans such as tensegrity robots (i; 26 robot built by s. Fivat), soft robots (j32), modular robots (k39) and robot swarms (l33).

(a) Hornby and Pollack12 (b) Auberbach and Bongard1 (a) hornby and Pollack12 (a) Hornby and Pollack12

Bongard et al. 4 (f) Loos et al. 16

(b) Auberbach and Bongard1

(b) auberbach and Bongard1

(a) Hornby and Pollack12 (b) Auberbach and Bongard1 The GOLEM Project Lipson and Pollack18

(c) (d)

(a) Hornby and Pollack12 (b) Auberbach and Bongard1 (a) Hornby and Pollack12 (b) Auberbach and Bongard1

y and Pollack12 (b) Auberbach and Bongard1 The GOLEM Project
Lipson and Pollack18
(c) (d)

(c) (a) Hornby and Pollack12 (b) Auberbach and Bongard1 The GOLEM Project Lipson and Pollack18

(c) (d)

(d) the GolEm Project lipson and Pollack18

(c) (c)

(g) Tuci et
rnby and Pollack12
and Pollack12

(b) Auberbach and Bongard1 (b) Auberbach and Bongard1

Bongard et al. 4 (f) Loos et al. 16 The GOLEM Project
(c) (d)
The GOLEM Project
Lipson and Pollack18

(c) (d)

(g) Tuci et a Bongard et al. 4 (f) Loos et al. 16 (g) Tuci et al. 35 (h) Hauert et al (e) Bongard et al. 4

nd Pollack12 (b) Auberbach and Bongard1 ngard et al. 4 (f) Loos et al. 16 Bongard et al. 4 (f) Loos et al. 16

Bongard et al. 4 ngard et al. 4

(f) Loos et al. 16 (f) Loos et al. 16

(c) (d)

The GOLEM Project
(i) Paul et al. 26 (j) Rieffel et al. 32 (k) Yi
(USC Polymo
(g) Tuci et al. 35 (h) Hauert et al. 11 (g) Tuci et al. 35 (h) Hauert et al. 11

(i) Paul et al. 26 (j) Rieffel et al. 32 (k) Yim et Lipson and Pollack18

 

Bongard et al. 4

(i) Paul et al. 26 (i) Paul et al. 26

(h) hauert et al. 11 (f) Loos et al. 16 (g) Tuci et al. 35 (h) Hauert et al Figure 1c, d, f, g,j,k,l reprinted courtesy of IEEE.

(j) Rieffel et al. 32 (k) Yim et al. 39

(l) Rubentstein e Figure 1e reprinted courtesy of AAAS.

(j) Rieffel et al. 32 (k) Yim et a (USC Polymorphic Robotic Lab)

Figure 1h reprinted courtesy of Springer.

(g) Tuci et al. 35 Bongard et al. 4

(g) tuci et al. 35

(g) Tuci et al. 35

(h) Hauert et al. 11

(h) Hauert et al. 11

Figure 1c,d,f,g,j,k,l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

Figure 1c,d,f,g,j,k,l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

Paul et al. 26 (i) Paul et al. 26

ard et al. 4

(j) Rieffel et al. 32 (k) Yim et al. 39

(USC Polymorphic Robotic Lab)

(l) Rubentstein et al. 33

(i) Paul et al. 26 (j) Rieffel et al. 32 (k) Yim et al. 39

(USC Polymorphic Robotic Lab)

(l) Rubentstein Figure 1c,d,f,g,j,k,l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

(i) Paul et al. 2 (i) Paul et al. 26 (j) Rieffel et al. 32 (k) Yi Figure 1c, d, f, g,j,k,l reprinted courtesy of IE Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer (g) Tuci et al. 35 (h) Hauert et al. 11

(j) Rieffel et al. 32 (k) Yim et al. 39 (USC Polymorphic Robotic Lab) (l) Rubentstein et al. 33

Figure 1c,d,f,g,j,k,l reprinted courtesy of I Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springe (USC Polymorphic Robotic Lab)

(l) Rubentstein et al. 33

Figure 1c, d,f,g,j, k,l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

(j) Rieffel et al. 32 (k) Yim et al. 39

(USC Polymorphic Robotic Lab)

(l) Rubentstein et al. 33

Figure 1c,d,f,g,j,k, l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

(f) Loos et al. 16

(i) Paul et al. 26 Paul et al. 26

ul et al. 26

figure 1c,d, f,g, j, k,l reprinted courtesy of iEEE. figure 1e reprinted courtesy of aaas. figure 1h reprinted courtesy of springer.

Figure 1c,d,f,g,j,k,l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

(j) Rieffel et al. 32

Figure 1c,d,f,g,j,k,l reprinted courtesy of IEEE.

Figure 1e reprinted courtesy of AAAS.

Figure 1h reprinted courtesy of Springer.

(k) Yim et al. 39

(USC Polymorphic Robotic Lab) (k) yim et al. 39

(usc Polymorphic Robotic lab)

(l) Rubentstein et al. 33 (l) Rubentstein et al. 33