Inspired by animals' ability to learn and adapt to changes in their environment during life, hybrid evolutionary algorithms have been developed and successfully applied in a number of research areas. This paper explores the effects of learning combined with a genetic algorithm to evolve control system parameters for a four-legged robot. Here, learning corresponds to the application of a local search algorithm on individuals during evolution. Two types of learning were implemented and tested, i.e. Baldwinian and Lamarckian learning. On the direct results from evolution in simulation, Lamarckian learning showed promising results, with a significant increase in final fitness compared with the results from evolution without learning. Further experiments with learning on the real robot demonstrated an efficient adaptation of the robot gait to the real world environment, and increased the performance to the level measured in simulation. This paper demonstrates that Lamarckian evolution is effective in improving the performance of robot controller evolution, and that the same learning process on the physical robot efficiently reduces the negative impact of the simulation-reality gap.

No abstract

No abstract

This paper presents an algorithm that makes an underactuated marine vehicle follow a straight line path while in the presence of a constant ocean current. When following the path, the vehicle maintains a desired surge speed which is measured relative to the water, and which may be constant or time-varying. The algorithm is an integral line-of-sight guidance law where the lookahead distance is designed to depend linearly on the desired relative surge speed of the vehicle. This dependency makes it possible to keep the maneuvering demands of the vehicle limited, even when the vehicle surge speed is large. It is shown that if the desired relative surge speed is constant along the path, the resulting error dynamics has a uniformly semiglobally exponentially stable equilibrium at the origin, thus achieving the path following and velocity control objectives. Furthermore, in the case of a general, timevarying desired speed trajectory, it is shown that the solutions of the system remain bounded. The results are supported by simulations, as well as experiments with an unmanned surface vehicle.

No abstract